diff --git "a/checkpoint-3052726/trainer_state.json" "b/checkpoint-3052726/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-3052726/trainer_state.json"
@@ -0,0 +1,36117 @@
+{
+  "best_metric": 4.024359226226807,
+  "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/outputs/pp-mod-subj2/lstm/1/checkpoints/checkpoint-3052726",
+  "epoch": 2.0249763653862156,
+  "eval_steps": 10,
+  "global_step": 3052726,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999998362119627e-05,
+      "loss": 10.8202,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999161405248948e-05,
+      "loss": 7.565,
+      "step": 512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.998322810497896e-05,
+      "loss": 7.0571,
+      "step": 1024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.997484215746844e-05,
+      "loss": 6.9897,
+      "step": 1536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.996645620995792e-05,
+      "loss": 6.9397,
+      "step": 2048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99580702624474e-05,
+      "loss": 6.8825,
+      "step": 2560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.994968431493688e-05,
+      "loss": 6.7131,
+      "step": 3072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.994129836742636e-05,
+      "loss": 6.6061,
+      "step": 3584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.993291241991584e-05,
+      "loss": 6.5156,
+      "step": 4096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.992454285120906e-05,
+      "loss": 6.4421,
+      "step": 4608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.991615690369854e-05,
+      "loss": 6.3918,
+      "step": 5120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.990777095618801e-05,
+      "loss": 6.3225,
+      "step": 5632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.989938500867749e-05,
+      "loss": 6.265,
+      "step": 6144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.989099906116697e-05,
+      "loss": 6.1905,
+      "step": 6656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.988262949246018e-05,
+      "loss": 6.1384,
+      "step": 7168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.987424354494966e-05,
+      "loss": 6.0719,
+      "step": 7680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.986585759743914e-05,
+      "loss": 6.0455,
+      "step": 8192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.985747164992862e-05,
+      "loss": 5.99,
+      "step": 8704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.98490857024181e-05,
+      "loss": 5.9458,
+      "step": 9216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.984069975490758e-05,
+      "loss": 5.9141,
+      "step": 9728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.983231380739706e-05,
+      "loss": 5.8881,
+      "step": 10240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9823927859886547e-05,
+      "loss": 5.8432,
+      "step": 10752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9815541912376026e-05,
+      "loss": 5.8186,
+      "step": 11264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9807155964865506e-05,
+      "loss": 5.782,
+      "step": 11776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9798770017354986e-05,
+      "loss": 5.7492,
+      "step": 12288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9790384069844466e-05,
+      "loss": 5.7304,
+      "step": 12800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9781998122333946e-05,
+      "loss": 5.6916,
+      "step": 13312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9773628553627155e-05,
+      "loss": 5.6783,
+      "step": 13824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9765242606116635e-05,
+      "loss": 5.6522,
+      "step": 14336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9756856658606115e-05,
+      "loss": 5.6232,
+      "step": 14848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9748470711095595e-05,
+      "loss": 5.6078,
+      "step": 15360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.974008476358507e-05,
+      "loss": 5.5848,
+      "step": 15872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.973169881607455e-05,
+      "loss": 5.5716,
+      "step": 16384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.972331286856403e-05,
+      "loss": 5.5522,
+      "step": 16896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9714926921053515e-05,
+      "loss": 5.5316,
+      "step": 17408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9706557352346724e-05,
+      "loss": 5.5232,
+      "step": 17920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9698171404836204e-05,
+      "loss": 5.5154,
+      "step": 18432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9689785457325684e-05,
+      "loss": 5.4763,
+      "step": 18944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9681399509815164e-05,
+      "loss": 5.4665,
+      "step": 19456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.967302994110837e-05,
+      "loss": 5.4527,
+      "step": 19968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.966464399359785e-05,
+      "loss": 5.4367,
+      "step": 20480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.965625804608733e-05,
+      "loss": 5.4215,
+      "step": 20992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.964787209857681e-05,
+      "loss": 5.4232,
+      "step": 21504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.963950252987002e-05,
+      "loss": 5.4026,
+      "step": 22016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.96311165823595e-05,
+      "loss": 5.3859,
+      "step": 22528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.962273063484898e-05,
+      "loss": 5.3791,
+      "step": 23040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.961434468733847e-05,
+      "loss": 5.369,
+      "step": 23552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.960597511863168e-05,
+      "loss": 5.3614,
+      "step": 24064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.959758917112116e-05,
+      "loss": 5.3365,
+      "step": 24576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9589219602414374e-05,
+      "loss": 5.3246,
+      "step": 25088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.958083365490385e-05,
+      "loss": 5.3328,
+      "step": 25600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.957244770739333e-05,
+      "loss": 5.3213,
+      "step": 26112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.956406175988281e-05,
+      "loss": 5.3086,
+      "step": 26624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.955567581237229e-05,
+      "loss": 5.2849,
+      "step": 27136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.954728986486177e-05,
+      "loss": 5.2934,
+      "step": 27648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.953890391735125e-05,
+      "loss": 5.2718,
+      "step": 28160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9530517969840727e-05,
+      "loss": 5.2871,
+      "step": 28672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9522148401133936e-05,
+      "loss": 5.2531,
+      "step": 29184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.951376245362342e-05,
+      "loss": 5.2562,
+      "step": 29696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.95053765061129e-05,
+      "loss": 5.2374,
+      "step": 30208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.949699055860238e-05,
+      "loss": 5.2245,
+      "step": 30720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.948862098989559e-05,
+      "loss": 5.2271,
+      "step": 31232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.94802514211888e-05,
+      "loss": 5.2106,
+      "step": 31744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.947186547367828e-05,
+      "loss": 5.2061,
+      "step": 32256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.946349590497149e-05,
+      "loss": 5.1972,
+      "step": 32768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.945510995746097e-05,
+      "loss": 5.208,
+      "step": 33280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.944672400995045e-05,
+      "loss": 5.1793,
+      "step": 33792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.943833806243993e-05,
+      "loss": 5.1684,
+      "step": 34304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.942995211492941e-05,
+      "loss": 5.1538,
+      "step": 34816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.942156616741889e-05,
+      "loss": 5.1533,
+      "step": 35328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.941318021990837e-05,
+      "loss": 5.1682,
+      "step": 35840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9404794272397856e-05,
+      "loss": 5.1576,
+      "step": 36352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9396408324887336e-05,
+      "loss": 5.1378,
+      "step": 36864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9388038756180545e-05,
+      "loss": 5.1343,
+      "step": 37376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9379669187473754e-05,
+      "loss": 5.1359,
+      "step": 37888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9371283239963234e-05,
+      "loss": 5.1313,
+      "step": 38400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9362897292452714e-05,
+      "loss": 5.1064,
+      "step": 38912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9354511344942194e-05,
+      "loss": 5.1101,
+      "step": 39424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9346125397431674e-05,
+      "loss": 5.0959,
+      "step": 39936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9337739449921154e-05,
+      "loss": 5.0987,
+      "step": 40448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9329353502410634e-05,
+      "loss": 5.0932,
+      "step": 40960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9320967554900114e-05,
+      "loss": 5.0831,
+      "step": 41472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9312581607389594e-05,
+      "loss": 5.0845,
+      "step": 41984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9304195659879074e-05,
+      "loss": 5.0742,
+      "step": 42496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.929582609117229e-05,
+      "loss": 5.0592,
+      "step": 43008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.928744014366177e-05,
+      "loss": 5.0584,
+      "step": 43520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.927905419615125e-05,
+      "loss": 5.0625,
+      "step": 44032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.927066824864073e-05,
+      "loss": 5.0564,
+      "step": 44544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.926229867993394e-05,
+      "loss": 5.044,
+      "step": 45056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.925391273242342e-05,
+      "loss": 5.0295,
+      "step": 45568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.92455267849129e-05,
+      "loss": 5.034,
+      "step": 46080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.923714083740238e-05,
+      "loss": 5.0238,
+      "step": 46592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.922875488989185e-05,
+      "loss": 5.0321,
+      "step": 47104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.922038532118507e-05,
+      "loss": 5.0103,
+      "step": 47616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.921199937367455e-05,
+      "loss": 5.0052,
+      "step": 48128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.920361342616403e-05,
+      "loss": 5.0222,
+      "step": 48640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.919522747865351e-05,
+      "loss": 4.9918,
+      "step": 49152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.918685790994672e-05,
+      "loss": 4.9859,
+      "step": 49664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.91784719624362e-05,
+      "loss": 4.9903,
+      "step": 50176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9170086014925676e-05,
+      "loss": 4.9879,
+      "step": 50688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9161700067415156e-05,
+      "loss": 4.9886,
+      "step": 51200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9153314119904636e-05,
+      "loss": 4.9756,
+      "step": 51712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.914494455119785e-05,
+      "loss": 4.9689,
+      "step": 52224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.913657498249106e-05,
+      "loss": 4.9619,
+      "step": 52736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.912818903498054e-05,
+      "loss": 4.9615,
+      "step": 53248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.911980308747002e-05,
+      "loss": 4.9464,
+      "step": 53760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.91114171399595e-05,
+      "loss": 4.9444,
+      "step": 54272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.910303119244898e-05,
+      "loss": 4.9507,
+      "step": 54784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.909464524493846e-05,
+      "loss": 4.9371,
+      "step": 55296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.908625929742794e-05,
+      "loss": 4.9313,
+      "step": 55808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.907788972872115e-05,
+      "loss": 4.9265,
+      "step": 56320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.906950378121063e-05,
+      "loss": 4.9172,
+      "step": 56832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.906111783370011e-05,
+      "loss": 4.9262,
+      "step": 57344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.905273188618959e-05,
+      "loss": 4.9116,
+      "step": 57856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.904434593867907e-05,
+      "loss": 4.9092,
+      "step": 58368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.903595999116855e-05,
+      "loss": 4.9115,
+      "step": 58880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.902757404365803e-05,
+      "loss": 4.9048,
+      "step": 59392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.901918809614751e-05,
+      "loss": 4.8998,
+      "step": 59904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.901080214863699e-05,
+      "loss": 4.8989,
+      "step": 60416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.900241620112647e-05,
+      "loss": 4.8986,
+      "step": 60928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.899403025361595e-05,
+      "loss": 4.8965,
+      "step": 61440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8985660684909166e-05,
+      "loss": 4.877,
+      "step": 61952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8977274737398646e-05,
+      "loss": 4.8869,
+      "step": 62464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8968888789888125e-05,
+      "loss": 4.8785,
+      "step": 62976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8960502842377605e-05,
+      "loss": 4.8793,
+      "step": 63488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8952133273670815e-05,
+      "loss": 4.8608,
+      "step": 64000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8943747326160294e-05,
+      "loss": 4.8699,
+      "step": 64512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8935361378649774e-05,
+      "loss": 4.8551,
+      "step": 65024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8926975431139254e-05,
+      "loss": 4.8616,
+      "step": 65536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8918605862432463e-05,
+      "loss": 4.8384,
+      "step": 66048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8910219914921943e-05,
+      "loss": 4.8631,
+      "step": 66560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8901833967411423e-05,
+      "loss": 4.8643,
+      "step": 67072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.88934480199009e-05,
+      "loss": 4.8451,
+      "step": 67584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.888506207239038e-05,
+      "loss": 4.8327,
+      "step": 68096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.88766925036836e-05,
+      "loss": 4.8434,
+      "step": 68608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.886830655617308e-05,
+      "loss": 4.8376,
+      "step": 69120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.885992060866256e-05,
+      "loss": 4.8352,
+      "step": 69632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.885153466115204e-05,
+      "loss": 4.8287,
+      "step": 70144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.884314871364151e-05,
+      "loss": 4.8312,
+      "step": 70656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.883477914493473e-05,
+      "loss": 4.8227,
+      "step": 71168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.882639319742421e-05,
+      "loss": 4.8197,
+      "step": 71680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.881800724991369e-05,
+      "loss": 4.8031,
+      "step": 72192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.880962130240316e-05,
+      "loss": 4.8156,
+      "step": 72704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.880123535489264e-05,
+      "loss": 4.8119,
+      "step": 73216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.879284940738212e-05,
+      "loss": 4.8016,
+      "step": 73728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.878447983867534e-05,
+      "loss": 4.8143,
+      "step": 74240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.877609389116482e-05,
+      "loss": 4.7929,
+      "step": 74752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.87677079436543e-05,
+      "loss": 4.8025,
+      "step": 75264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.875932199614378e-05,
+      "loss": 4.8035,
+      "step": 75776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8750952427436986e-05,
+      "loss": 4.7926,
+      "step": 76288
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.804625034332275,
+      "eval_runtime": 301.7618,
+      "eval_samples_per_second": 1264.544,
+      "eval_steps_per_second": 39.518,
+      "step": 76320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8742566479926466e-05,
+      "loss": 4.7806,
+      "step": 76800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8734180532415946e-05,
+      "loss": 4.7868,
+      "step": 77312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8725794584905426e-05,
+      "loss": 4.7899,
+      "step": 77824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8717408637394906e-05,
+      "loss": 4.7802,
+      "step": 78336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8709022689884386e-05,
+      "loss": 4.7817,
+      "step": 78848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8700636742373866e-05,
+      "loss": 4.7649,
+      "step": 79360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8692267173667075e-05,
+      "loss": 4.7694,
+      "step": 79872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8683881226156555e-05,
+      "loss": 4.7599,
+      "step": 80384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.867549527864604e-05,
+      "loss": 4.766,
+      "step": 80896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.866710933113552e-05,
+      "loss": 4.7559,
+      "step": 81408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.865873976242873e-05,
+      "loss": 4.7621,
+      "step": 81920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.865035381491821e-05,
+      "loss": 4.7635,
+      "step": 82432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.864198424621142e-05,
+      "loss": 4.7489,
+      "step": 82944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.86335982987009e-05,
+      "loss": 4.7473,
+      "step": 83456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.862521235119038e-05,
+      "loss": 4.7321,
+      "step": 83968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.861682640367986e-05,
+      "loss": 4.7431,
+      "step": 84480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.860844045616934e-05,
+      "loss": 4.738,
+      "step": 84992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.860005450865882e-05,
+      "loss": 4.7312,
+      "step": 85504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.85916685611483e-05,
+      "loss": 4.7289,
+      "step": 86016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.858328261363778e-05,
+      "loss": 4.756,
+      "step": 86528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8574913044930995e-05,
+      "loss": 4.7278,
+      "step": 87040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8566527097420475e-05,
+      "loss": 4.7296,
+      "step": 87552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8558141149909955e-05,
+      "loss": 4.7316,
+      "step": 88064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8549755202399435e-05,
+      "loss": 4.7229,
+      "step": 88576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8541385633692644e-05,
+      "loss": 4.7204,
+      "step": 89088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8532999686182124e-05,
+      "loss": 4.7083,
+      "step": 89600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8524613738671604e-05,
+      "loss": 4.7184,
+      "step": 90112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8516227791161084e-05,
+      "loss": 4.713,
+      "step": 90624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8507841843650564e-05,
+      "loss": 4.699,
+      "step": 91136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.849947227494377e-05,
+      "loss": 4.7083,
+      "step": 91648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.849108632743325e-05,
+      "loss": 4.6985,
+      "step": 92160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.848270037992273e-05,
+      "loss": 4.7074,
+      "step": 92672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.847431443241221e-05,
+      "loss": 4.7058,
+      "step": 93184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.846592848490169e-05,
+      "loss": 4.6922,
+      "step": 93696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.845754253739117e-05,
+      "loss": 4.7009,
+      "step": 94208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.844915658988065e-05,
+      "loss": 4.7083,
+      "step": 94720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.844077064237013e-05,
+      "loss": 4.6775,
+      "step": 95232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.843238469485961e-05,
+      "loss": 4.688,
+      "step": 95744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.842399874734909e-05,
+      "loss": 4.6853,
+      "step": 96256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.841561279983857e-05,
+      "loss": 4.6743,
+      "step": 96768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.840722685232805e-05,
+      "loss": 4.6785,
+      "step": 97280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.839885728362126e-05,
+      "loss": 4.6872,
+      "step": 97792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.839047133611074e-05,
+      "loss": 4.6794,
+      "step": 98304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.838210176740395e-05,
+      "loss": 4.674,
+      "step": 98816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.837371581989343e-05,
+      "loss": 4.6777,
+      "step": 99328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.836532987238292e-05,
+      "loss": 4.6713,
+      "step": 99840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.83569439248724e-05,
+      "loss": 4.6739,
+      "step": 100352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.834855797736188e-05,
+      "loss": 4.6567,
+      "step": 100864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.834020478745882e-05,
+      "loss": 4.6537,
+      "step": 101376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8331818839948295e-05,
+      "loss": 4.6705,
+      "step": 101888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8323432892437775e-05,
+      "loss": 4.6685,
+      "step": 102400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8315046944927255e-05,
+      "loss": 4.6573,
+      "step": 102912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8306660997416735e-05,
+      "loss": 4.648,
+      "step": 103424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8298275049906215e-05,
+      "loss": 4.6585,
+      "step": 103936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8289889102395695e-05,
+      "loss": 4.6407,
+      "step": 104448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8281503154885175e-05,
+      "loss": 4.6655,
+      "step": 104960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8273117207374655e-05,
+      "loss": 4.6444,
+      "step": 105472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8264731259864135e-05,
+      "loss": 4.6551,
+      "step": 105984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.825636169115735e-05,
+      "loss": 4.6397,
+      "step": 106496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.824797574364683e-05,
+      "loss": 4.6339,
+      "step": 107008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.823958979613631e-05,
+      "loss": 4.6359,
+      "step": 107520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.823120384862579e-05,
+      "loss": 4.6421,
+      "step": 108032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8222834279919e-05,
+      "loss": 4.6251,
+      "step": 108544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.821446471121221e-05,
+      "loss": 4.6283,
+      "step": 109056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.820609514250542e-05,
+      "loss": 4.643,
+      "step": 109568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.81977091949949e-05,
+      "loss": 4.6237,
+      "step": 110080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.818932324748438e-05,
+      "loss": 4.6142,
+      "step": 110592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.818093729997386e-05,
+      "loss": 4.6164,
+      "step": 111104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.817255135246334e-05,
+      "loss": 4.6138,
+      "step": 111616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.816416540495282e-05,
+      "loss": 4.6342,
+      "step": 112128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8155779457442305e-05,
+      "loss": 4.629,
+      "step": 112640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8147393509931785e-05,
+      "loss": 4.6125,
+      "step": 113152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8139007562421265e-05,
+      "loss": 4.6176,
+      "step": 113664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8130621614910745e-05,
+      "loss": 4.6209,
+      "step": 114176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8122235667400224e-05,
+      "loss": 4.619,
+      "step": 114688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8113849719889704e-05,
+      "loss": 4.6041,
+      "step": 115200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8105463772379184e-05,
+      "loss": 4.6155,
+      "step": 115712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.809707782486866e-05,
+      "loss": 4.6029,
+      "step": 116224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8088708256161873e-05,
+      "loss": 4.6104,
+      "step": 116736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.808032230865135e-05,
+      "loss": 4.6027,
+      "step": 117248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8071936361140827e-05,
+      "loss": 4.6037,
+      "step": 117760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8063550413630307e-05,
+      "loss": 4.6058,
+      "step": 118272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8055164466119786e-05,
+      "loss": 4.6065,
+      "step": 118784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.804677851860927e-05,
+      "loss": 4.588,
+      "step": 119296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.803839257109875e-05,
+      "loss": 4.5969,
+      "step": 119808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.803000662358823e-05,
+      "loss": 4.5943,
+      "step": 120320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.802163705488144e-05,
+      "loss": 4.5994,
+      "step": 120832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.801325110737092e-05,
+      "loss": 4.5933,
+      "step": 121344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.80048651598604e-05,
+      "loss": 4.5805,
+      "step": 121856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.799647921234988e-05,
+      "loss": 4.5894,
+      "step": 122368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.798812602244683e-05,
+      "loss": 4.5829,
+      "step": 122880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.79797400749363e-05,
+      "loss": 4.5951,
+      "step": 123392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.797135412742578e-05,
+      "loss": 4.5734,
+      "step": 123904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.796296817991526e-05,
+      "loss": 4.5781,
+      "step": 124416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7954598611208476e-05,
+      "loss": 4.5907,
+      "step": 124928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7946212663697956e-05,
+      "loss": 4.58,
+      "step": 125440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7937826716187436e-05,
+      "loss": 4.5609,
+      "step": 125952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7929440768676916e-05,
+      "loss": 4.5733,
+      "step": 126464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7921054821166396e-05,
+      "loss": 4.5782,
+      "step": 126976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7912668873655876e-05,
+      "loss": 4.5794,
+      "step": 127488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7904282926145356e-05,
+      "loss": 4.5689,
+      "step": 128000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7895896978634836e-05,
+      "loss": 4.5665,
+      "step": 128512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7887527409928045e-05,
+      "loss": 4.5613,
+      "step": 129024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7879141462417525e-05,
+      "loss": 4.5689,
+      "step": 129536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7870755514907005e-05,
+      "loss": 4.5495,
+      "step": 130048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7862385946200214e-05,
+      "loss": 4.5571,
+      "step": 130560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7853999998689694e-05,
+      "loss": 4.5642,
+      "step": 131072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.784561405117918e-05,
+      "loss": 4.5616,
+      "step": 131584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.783724448247239e-05,
+      "loss": 4.5444,
+      "step": 132096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.782885853496187e-05,
+      "loss": 4.5549,
+      "step": 132608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.782047258745135e-05,
+      "loss": 4.5428,
+      "step": 133120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.781208663994083e-05,
+      "loss": 4.5575,
+      "step": 133632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.780370069243031e-05,
+      "loss": 4.5457,
+      "step": 134144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.779531474491979e-05,
+      "loss": 4.5361,
+      "step": 134656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.778692879740927e-05,
+      "loss": 4.5559,
+      "step": 135168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.777854284989875e-05,
+      "loss": 4.5483,
+      "step": 135680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.777015690238823e-05,
+      "loss": 4.5431,
+      "step": 136192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.776177095487771e-05,
+      "loss": 4.546,
+      "step": 136704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.775338500736719e-05,
+      "loss": 4.544,
+      "step": 137216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.774499905985666e-05,
+      "loss": 4.5477,
+      "step": 137728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.773661311234615e-05,
+      "loss": 4.528,
+      "step": 138240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.772822716483563e-05,
+      "loss": 4.5473,
+      "step": 138752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.771984121732511e-05,
+      "loss": 4.534,
+      "step": 139264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.771145526981459e-05,
+      "loss": 4.5389,
+      "step": 139776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.77030857011078e-05,
+      "loss": 4.5288,
+      "step": 140288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.769469975359728e-05,
+      "loss": 4.5355,
+      "step": 140800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.768631380608676e-05,
+      "loss": 4.5233,
+      "step": 141312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.767792785857624e-05,
+      "loss": 4.5317,
+      "step": 141824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.766955828986945e-05,
+      "loss": 4.5102,
+      "step": 142336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.766117234235893e-05,
+      "loss": 4.5418,
+      "step": 142848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.765278639484841e-05,
+      "loss": 4.536,
+      "step": 143360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.764440044733789e-05,
+      "loss": 4.528,
+      "step": 143872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7636030878631096e-05,
+      "loss": 4.5158,
+      "step": 144384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.762764493112058e-05,
+      "loss": 4.53,
+      "step": 144896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.761925898361006e-05,
+      "loss": 4.5308,
+      "step": 145408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.761087303609954e-05,
+      "loss": 4.5215,
+      "step": 145920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.760250346739275e-05,
+      "loss": 4.5158,
+      "step": 146432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.759411751988223e-05,
+      "loss": 4.5248,
+      "step": 146944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.758573157237171e-05,
+      "loss": 4.5194,
+      "step": 147456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.757734562486119e-05,
+      "loss": 4.5255,
+      "step": 147968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.75689760561544e-05,
+      "loss": 4.5012,
+      "step": 148480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.756059010864388e-05,
+      "loss": 4.5175,
+      "step": 148992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.755220416113336e-05,
+      "loss": 4.5151,
+      "step": 149504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.754381821362284e-05,
+      "loss": 4.5082,
+      "step": 150016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.753544864491605e-05,
+      "loss": 4.523,
+      "step": 150528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7527062697405536e-05,
+      "loss": 4.5019,
+      "step": 151040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7518676749895016e-05,
+      "loss": 4.5123,
+      "step": 151552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7510290802384496e-05,
+      "loss": 4.5087,
+      "step": 152064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7501904854873976e-05,
+      "loss": 4.5099,
+      "step": 152576
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.525796890258789,
+      "eval_runtime": 294.223,
+      "eval_samples_per_second": 1296.945,
+      "eval_steps_per_second": 40.53,
+      "step": 152640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7493518907363456e-05,
+      "loss": 4.4976,
+      "step": 153088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7485132959852936e-05,
+      "loss": 4.5046,
+      "step": 153600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7476747012342416e-05,
+      "loss": 4.5098,
+      "step": 154112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7468361064831896e-05,
+      "loss": 4.5045,
+      "step": 154624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7459975117321376e-05,
+      "loss": 4.5045,
+      "step": 155136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.745158916981085e-05,
+      "loss": 4.4881,
+      "step": 155648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.744320322230033e-05,
+      "loss": 4.4964,
+      "step": 156160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.743481727478981e-05,
+      "loss": 4.4891,
+      "step": 156672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.742643132727929e-05,
+      "loss": 4.4999,
+      "step": 157184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.741804537976877e-05,
+      "loss": 4.489,
+      "step": 157696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.740965943225825e-05,
+      "loss": 4.4994,
+      "step": 158208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7401273484747736e-05,
+      "loss": 4.5003,
+      "step": 158720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7392903916040945e-05,
+      "loss": 4.4829,
+      "step": 159232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7384517968530425e-05,
+      "loss": 4.4888,
+      "step": 159744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7376132021019905e-05,
+      "loss": 4.4709,
+      "step": 160256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7367746073509385e-05,
+      "loss": 4.4824,
+      "step": 160768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7359360125998865e-05,
+      "loss": 4.4835,
+      "step": 161280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7350974178488345e-05,
+      "loss": 4.4768,
+      "step": 161792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7342588230977825e-05,
+      "loss": 4.4762,
+      "step": 162304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7334202283467305e-05,
+      "loss": 4.501,
+      "step": 162816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7325816335956784e-05,
+      "loss": 4.4794,
+      "step": 163328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7317430388446264e-05,
+      "loss": 4.4766,
+      "step": 163840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.730904444093574e-05,
+      "loss": 4.4832,
+      "step": 164352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7300674872228953e-05,
+      "loss": 4.4809,
+      "step": 164864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7292288924718433e-05,
+      "loss": 4.4731,
+      "step": 165376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.728390297720791e-05,
+      "loss": 4.4638,
+      "step": 165888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.727553340850113e-05,
+      "loss": 4.4749,
+      "step": 166400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.726714746099061e-05,
+      "loss": 4.4687,
+      "step": 166912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.725876151348009e-05,
+      "loss": 4.4612,
+      "step": 167424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.725037556596957e-05,
+      "loss": 4.4712,
+      "step": 167936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.724198961845904e-05,
+      "loss": 4.4609,
+      "step": 168448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.723360367094852e-05,
+      "loss": 4.4709,
+      "step": 168960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.722523410224174e-05,
+      "loss": 4.4682,
+      "step": 169472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.721684815473121e-05,
+      "loss": 4.4588,
+      "step": 169984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.720847858602443e-05,
+      "loss": 4.4705,
+      "step": 170496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.720009263851391e-05,
+      "loss": 4.4812,
+      "step": 171008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.719170669100339e-05,
+      "loss": 4.4471,
+      "step": 171520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.718332074349287e-05,
+      "loss": 4.4633,
+      "step": 172032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.717493479598235e-05,
+      "loss": 4.4523,
+      "step": 172544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.716654884847183e-05,
+      "loss": 4.4548,
+      "step": 173056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.715816290096131e-05,
+      "loss": 4.4509,
+      "step": 173568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.714977695345079e-05,
+      "loss": 4.4682,
+      "step": 174080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.714139100594027e-05,
+      "loss": 4.4584,
+      "step": 174592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.713300505842975e-05,
+      "loss": 4.4537,
+      "step": 175104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7124635489722956e-05,
+      "loss": 4.456,
+      "step": 175616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7116249542212436e-05,
+      "loss": 4.449,
+      "step": 176128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7107863594701916e-05,
+      "loss": 4.4614,
+      "step": 176640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7099477647191396e-05,
+      "loss": 4.4409,
+      "step": 177152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7091091699680876e-05,
+      "loss": 4.4385,
+      "step": 177664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7082705752170356e-05,
+      "loss": 4.4541,
+      "step": 178176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7074319804659836e-05,
+      "loss": 4.4526,
+      "step": 178688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.706593385714932e-05,
+      "loss": 4.4436,
+      "step": 179200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.705758066724626e-05,
+      "loss": 4.4395,
+      "step": 179712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.704919471973574e-05,
+      "loss": 4.4481,
+      "step": 180224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.704080877222522e-05,
+      "loss": 4.4313,
+      "step": 180736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.70324228247147e-05,
+      "loss": 4.4558,
+      "step": 181248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.702403687720418e-05,
+      "loss": 4.4389,
+      "step": 181760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.701565092969366e-05,
+      "loss": 4.4491,
+      "step": 182272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.700726498218314e-05,
+      "loss": 4.4428,
+      "step": 182784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.699889541347635e-05,
+      "loss": 4.4312,
+      "step": 183296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.699050946596583e-05,
+      "loss": 4.4311,
+      "step": 183808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.698212351845531e-05,
+      "loss": 4.4417,
+      "step": 184320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.697373757094479e-05,
+      "loss": 4.4216,
+      "step": 184832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6965351623434276e-05,
+      "loss": 4.4294,
+      "step": 185344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6956965675923756e-05,
+      "loss": 4.4402,
+      "step": 185856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.694857972841323e-05,
+      "loss": 4.4304,
+      "step": 186368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.694019378090271e-05,
+      "loss": 4.4173,
+      "step": 186880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6931824212195925e-05,
+      "loss": 4.419,
+      "step": 187392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.69234382646854e-05,
+      "loss": 4.4174,
+      "step": 187904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.691505231717488e-05,
+      "loss": 4.4429,
+      "step": 188416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.690666636966436e-05,
+      "loss": 4.4321,
+      "step": 188928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6898296800957574e-05,
+      "loss": 4.4227,
+      "step": 189440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.688991085344705e-05,
+      "loss": 4.4299,
+      "step": 189952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.688152490593653e-05,
+      "loss": 4.4255,
+      "step": 190464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6873138958426014e-05,
+      "loss": 4.4302,
+      "step": 190976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.686478576852296e-05,
+      "loss": 4.4141,
+      "step": 191488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.685639982101244e-05,
+      "loss": 4.4284,
+      "step": 192000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.684801387350192e-05,
+      "loss": 4.4211,
+      "step": 192512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.68396279259914e-05,
+      "loss": 4.4203,
+      "step": 193024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.683124197848087e-05,
+      "loss": 4.4158,
+      "step": 193536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.682285603097035e-05,
+      "loss": 4.4197,
+      "step": 194048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.681447008345983e-05,
+      "loss": 4.42,
+      "step": 194560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.680610051475305e-05,
+      "loss": 4.4235,
+      "step": 195072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.679771456724252e-05,
+      "loss": 4.4094,
+      "step": 195584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6789328619732e-05,
+      "loss": 4.4157,
+      "step": 196096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.678094267222148e-05,
+      "loss": 4.4106,
+      "step": 196608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.677255672471097e-05,
+      "loss": 4.4166,
+      "step": 197120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6764187156004177e-05,
+      "loss": 4.4165,
+      "step": 197632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6755801208493657e-05,
+      "loss": 4.3979,
+      "step": 198144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6747415260983136e-05,
+      "loss": 4.4119,
+      "step": 198656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6739029313472616e-05,
+      "loss": 4.4071,
+      "step": 199168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6730659744765826e-05,
+      "loss": 4.4179,
+      "step": 199680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6722273797255305e-05,
+      "loss": 4.3948,
+      "step": 200192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6713887849744785e-05,
+      "loss": 4.4053,
+      "step": 200704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6705501902234265e-05,
+      "loss": 4.4168,
+      "step": 201216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6697115954723745e-05,
+      "loss": 4.4043,
+      "step": 201728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6688730007213225e-05,
+      "loss": 4.3887,
+      "step": 202240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6680344059702705e-05,
+      "loss": 4.4018,
+      "step": 202752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.667197449099592e-05,
+      "loss": 4.4077,
+      "step": 203264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.66635885434854e-05,
+      "loss": 4.4087,
+      "step": 203776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.665520259597488e-05,
+      "loss": 4.401,
+      "step": 204288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.664681664846436e-05,
+      "loss": 4.3935,
+      "step": 204800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.663843070095384e-05,
+      "loss": 4.3931,
+      "step": 205312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.663004475344332e-05,
+      "loss": 4.4001,
+      "step": 205824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.66216588059328e-05,
+      "loss": 4.3837,
+      "step": 206336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.661327285842228e-05,
+      "loss": 4.3909,
+      "step": 206848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.660490328971549e-05,
+      "loss": 4.398,
+      "step": 207360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.659651734220497e-05,
+      "loss": 4.3956,
+      "step": 207872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.658813139469445e-05,
+      "loss": 4.3779,
+      "step": 208384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.657974544718393e-05,
+      "loss": 4.3943,
+      "step": 208896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.657135949967341e-05,
+      "loss": 4.3805,
+      "step": 209408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.656297355216288e-05,
+      "loss": 4.3897,
+      "step": 209920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.655458760465237e-05,
+      "loss": 4.386,
+      "step": 210432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6546218035945586e-05,
+      "loss": 4.3776,
+      "step": 210944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6537832088435065e-05,
+      "loss": 4.3936,
+      "step": 211456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.652944614092454e-05,
+      "loss": 4.3912,
+      "step": 211968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.652106019341402e-05,
+      "loss": 4.3811,
+      "step": 212480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6512690624707234e-05,
+      "loss": 4.3871,
+      "step": 212992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.650430467719671e-05,
+      "loss": 4.3865,
+      "step": 213504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.649591872968619e-05,
+      "loss": 4.388,
+      "step": 214016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.648753278217567e-05,
+      "loss": 4.374,
+      "step": 214528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6479163213468883e-05,
+      "loss": 4.3874,
+      "step": 215040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.647077726595836e-05,
+      "loss": 4.3801,
+      "step": 215552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6462391318447837e-05,
+      "loss": 4.3813,
+      "step": 216064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.645400537093732e-05,
+      "loss": 4.3782,
+      "step": 216576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.64456194234268e-05,
+      "loss": 4.3852,
+      "step": 217088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.643723347591628e-05,
+      "loss": 4.3658,
+      "step": 217600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.642886390720949e-05,
+      "loss": 4.3779,
+      "step": 218112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.642047795969897e-05,
+      "loss": 4.3597,
+      "step": 218624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.641209201218845e-05,
+      "loss": 4.3873,
+      "step": 219136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.640370606467793e-05,
+      "loss": 4.3817,
+      "step": 219648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.639532011716741e-05,
+      "loss": 4.3781,
+      "step": 220160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.638693416965689e-05,
+      "loss": 4.3646,
+      "step": 220672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.637854822214637e-05,
+      "loss": 4.3784,
+      "step": 221184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.637016227463585e-05,
+      "loss": 4.3839,
+      "step": 221696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.636179270592906e-05,
+      "loss": 4.3694,
+      "step": 222208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.635340675841854e-05,
+      "loss": 4.3687,
+      "step": 222720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.634502081090802e-05,
+      "loss": 4.3768,
+      "step": 223232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.633663486339751e-05,
+      "loss": 4.3705,
+      "step": 223744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.632824891588699e-05,
+      "loss": 4.3785,
+      "step": 224256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.63198793471802e-05,
+      "loss": 4.3594,
+      "step": 224768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.631149339966968e-05,
+      "loss": 4.3701,
+      "step": 225280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.630310745215916e-05,
+      "loss": 4.369,
+      "step": 225792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.629472150464864e-05,
+      "loss": 4.3604,
+      "step": 226304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6286351935941846e-05,
+      "loss": 4.3807,
+      "step": 226816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6277965988431326e-05,
+      "loss": 4.358,
+      "step": 227328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6269580040920806e-05,
+      "loss": 4.3667,
+      "step": 227840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6261194093410286e-05,
+      "loss": 4.3684,
+      "step": 228352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6252824524703495e-05,
+      "loss": 4.3635,
+      "step": 228864
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.390058994293213,
+      "eval_runtime": 286.6071,
+      "eval_samples_per_second": 1331.408,
+      "eval_steps_per_second": 41.607,
+      "step": 228960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6244438577192975e-05,
+      "loss": 4.3622,
+      "step": 229376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.623605262968246e-05,
+      "loss": 4.3597,
+      "step": 229888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.622766668217194e-05,
+      "loss": 4.3701,
+      "step": 230400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.621929711346515e-05,
+      "loss": 4.3641,
+      "step": 230912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.621091116595463e-05,
+      "loss": 4.365,
+      "step": 231424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.620252521844411e-05,
+      "loss": 4.3519,
+      "step": 231936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.619413927093359e-05,
+      "loss": 4.359,
+      "step": 232448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.618575332342307e-05,
+      "loss": 4.3435,
+      "step": 232960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.617738375471628e-05,
+      "loss": 4.3675,
+      "step": 233472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.616899780720576e-05,
+      "loss": 4.3512,
+      "step": 233984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.616062823849897e-05,
+      "loss": 4.3615,
+      "step": 234496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.615224229098845e-05,
+      "loss": 4.3651,
+      "step": 235008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6143872722281664e-05,
+      "loss": 4.3518,
+      "step": 235520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6135486774771144e-05,
+      "loss": 4.3463,
+      "step": 236032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6127100827260624e-05,
+      "loss": 4.3435,
+      "step": 236544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6118714879750104e-05,
+      "loss": 4.3425,
+      "step": 237056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6110328932239584e-05,
+      "loss": 4.3533,
+      "step": 237568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6101942984729064e-05,
+      "loss": 4.3422,
+      "step": 238080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6093557037218544e-05,
+      "loss": 4.3437,
+      "step": 238592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.608517108970802e-05,
+      "loss": 4.3656,
+      "step": 239104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.60767851421975e-05,
+      "loss": 4.3486,
+      "step": 239616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.606839919468698e-05,
+      "loss": 4.3464,
+      "step": 240128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.606001324717646e-05,
+      "loss": 4.3537,
+      "step": 240640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.605162729966594e-05,
+      "loss": 4.3493,
+      "step": 241152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.604327410976288e-05,
+      "loss": 4.3438,
+      "step": 241664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.603488816225236e-05,
+      "loss": 4.3321,
+      "step": 242176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.602650221474184e-05,
+      "loss": 4.349,
+      "step": 242688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.601811626723132e-05,
+      "loss": 4.3315,
+      "step": 243200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.60097303197208e-05,
+      "loss": 4.3362,
+      "step": 243712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.600134437221028e-05,
+      "loss": 4.3384,
+      "step": 244224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.599295842469976e-05,
+      "loss": 4.3328,
+      "step": 244736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.598457247718924e-05,
+      "loss": 4.3467,
+      "step": 245248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.597620290848245e-05,
+      "loss": 4.3398,
+      "step": 245760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.596781696097193e-05,
+      "loss": 4.3318,
+      "step": 246272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.595944739226514e-05,
+      "loss": 4.3437,
+      "step": 246784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.595106144475462e-05,
+      "loss": 4.3492,
+      "step": 247296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.59426754972441e-05,
+      "loss": 4.3281,
+      "step": 247808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5934289549733587e-05,
+      "loss": 4.3386,
+      "step": 248320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5925903602223066e-05,
+      "loss": 4.3233,
+      "step": 248832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5917517654712546e-05,
+      "loss": 4.3398,
+      "step": 249344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5909131707202026e-05,
+      "loss": 4.3239,
+      "step": 249856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5900745759691506e-05,
+      "loss": 4.3439,
+      "step": 250368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5892376190984715e-05,
+      "loss": 4.3336,
+      "step": 250880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5883990243474195e-05,
+      "loss": 4.3316,
+      "step": 251392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5875604295963675e-05,
+      "loss": 4.3362,
+      "step": 251904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5867234727256884e-05,
+      "loss": 4.3254,
+      "step": 252416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5858865158550094e-05,
+      "loss": 4.3378,
+      "step": 252928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5850479211039573e-05,
+      "loss": 4.3227,
+      "step": 253440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5842093263529053e-05,
+      "loss": 4.3176,
+      "step": 253952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.583370731601854e-05,
+      "loss": 4.3353,
+      "step": 254464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.582532136850802e-05,
+      "loss": 4.3283,
+      "step": 254976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.58169354209975e-05,
+      "loss": 4.3293,
+      "step": 255488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.580854947348698e-05,
+      "loss": 4.3178,
+      "step": 256000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.580016352597646e-05,
+      "loss": 4.3271,
+      "step": 256512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.579177757846594e-05,
+      "loss": 4.3143,
+      "step": 257024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.578340800975915e-05,
+      "loss": 4.336,
+      "step": 257536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.577502206224863e-05,
+      "loss": 4.3208,
+      "step": 258048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.576663611473811e-05,
+      "loss": 4.3335,
+      "step": 258560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.575825016722759e-05,
+      "loss": 4.3263,
+      "step": 259072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.574986421971707e-05,
+      "loss": 4.3178,
+      "step": 259584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.574147827220655e-05,
+      "loss": 4.3122,
+      "step": 260096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.573309232469602e-05,
+      "loss": 4.3255,
+      "step": 260608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.572472275598924e-05,
+      "loss": 4.3081,
+      "step": 261120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5716336808478725e-05,
+      "loss": 4.3147,
+      "step": 261632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5707950860968205e-05,
+      "loss": 4.3201,
+      "step": 262144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.569956491345768e-05,
+      "loss": 4.3202,
+      "step": 262656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.569117896594716e-05,
+      "loss": 4.2991,
+      "step": 263168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5682809397240374e-05,
+      "loss": 4.3072,
+      "step": 263680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5674423449729854e-05,
+      "loss": 4.301,
+      "step": 264192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.566603750221933e-05,
+      "loss": 4.3307,
+      "step": 264704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.565765155470881e-05,
+      "loss": 4.3171,
+      "step": 265216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.564926560719829e-05,
+      "loss": 4.3121,
+      "step": 265728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5640879659687767e-05,
+      "loss": 4.3181,
+      "step": 266240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5632493712177247e-05,
+      "loss": 4.3114,
+      "step": 266752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5624107764666726e-05,
+      "loss": 4.3246,
+      "step": 267264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.561575457476368e-05,
+      "loss": 4.2993,
+      "step": 267776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.560736862725315e-05,
+      "loss": 4.3214,
+      "step": 268288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.559898267974263e-05,
+      "loss": 4.3068,
+      "step": 268800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.559059673223211e-05,
+      "loss": 4.3107,
+      "step": 269312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.558221078472159e-05,
+      "loss": 4.311,
+      "step": 269824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.557382483721107e-05,
+      "loss": 4.3065,
+      "step": 270336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.556543888970055e-05,
+      "loss": 4.3148,
+      "step": 270848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.555705294219003e-05,
+      "loss": 4.3104,
+      "step": 271360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.554868337348324e-05,
+      "loss": 4.3012,
+      "step": 271872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.554029742597272e-05,
+      "loss": 4.3095,
+      "step": 272384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.55319114784622e-05,
+      "loss": 4.2998,
+      "step": 272896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.552352553095168e-05,
+      "loss": 4.315,
+      "step": 273408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5515155962244896e-05,
+      "loss": 4.3071,
+      "step": 273920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5506770014734376e-05,
+      "loss": 4.2913,
+      "step": 274432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5498384067223856e-05,
+      "loss": 4.3068,
+      "step": 274944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5489998119713336e-05,
+      "loss": 4.2966,
+      "step": 275456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5481628551006545e-05,
+      "loss": 4.3083,
+      "step": 275968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5473242603496025e-05,
+      "loss": 4.2924,
+      "step": 276480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5464856655985505e-05,
+      "loss": 4.302,
+      "step": 276992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5456470708474985e-05,
+      "loss": 4.3095,
+      "step": 277504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5448084760964465e-05,
+      "loss": 4.2959,
+      "step": 278016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5439698813453945e-05,
+      "loss": 4.2894,
+      "step": 278528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5431312865943425e-05,
+      "loss": 4.2958,
+      "step": 279040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5422926918432905e-05,
+      "loss": 4.2958,
+      "step": 279552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5414557349726114e-05,
+      "loss": 4.31,
+      "step": 280064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5406171402215594e-05,
+      "loss": 4.2959,
+      "step": 280576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.539778545470508e-05,
+      "loss": 4.2915,
+      "step": 281088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.538941588599829e-05,
+      "loss": 4.2908,
+      "step": 281600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.538106269609523e-05,
+      "loss": 4.2999,
+      "step": 282112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.537267674858471e-05,
+      "loss": 4.2773,
+      "step": 282624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.536429080107419e-05,
+      "loss": 4.2916,
+      "step": 283136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.535590485356367e-05,
+      "loss": 4.2926,
+      "step": 283648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.534751890605315e-05,
+      "loss": 4.2952,
+      "step": 284160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.533913295854263e-05,
+      "loss": 4.2731,
+      "step": 284672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.533074701103211e-05,
+      "loss": 4.2922,
+      "step": 285184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.532236106352159e-05,
+      "loss": 4.2863,
+      "step": 285696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.531397511601107e-05,
+      "loss": 4.2818,
+      "step": 286208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.530558916850055e-05,
+      "loss": 4.2909,
+      "step": 286720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5297203220990034e-05,
+      "loss": 4.2736,
+      "step": 287232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5288817273479514e-05,
+      "loss": 4.2958,
+      "step": 287744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.528046408357645e-05,
+      "loss": 4.2898,
+      "step": 288256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.527207813606593e-05,
+      "loss": 4.2828,
+      "step": 288768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.526369218855541e-05,
+      "loss": 4.2893,
+      "step": 289280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.525530624104489e-05,
+      "loss": 4.2828,
+      "step": 289792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.524692029353437e-05,
+      "loss": 4.2905,
+      "step": 290304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.523853434602385e-05,
+      "loss": 4.2702,
+      "step": 290816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.523014839851333e-05,
+      "loss": 4.2945,
+      "step": 291328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5221762451002805e-05,
+      "loss": 4.2806,
+      "step": 291840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5213376503492285e-05,
+      "loss": 4.2812,
+      "step": 292352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.52050069347855e-05,
+      "loss": 4.2809,
+      "step": 292864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.519662098727499e-05,
+      "loss": 4.2881,
+      "step": 293376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.518823503976446e-05,
+      "loss": 4.2751,
+      "step": 293888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.517984909225394e-05,
+      "loss": 4.2767,
+      "step": 294400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.517147952354716e-05,
+      "loss": 4.2674,
+      "step": 294912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.516309357603663e-05,
+      "loss": 4.2898,
+      "step": 295424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.515470762852611e-05,
+      "loss": 4.2848,
+      "step": 295936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.514632168101559e-05,
+      "loss": 4.2829,
+      "step": 296448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5137952112308806e-05,
+      "loss": 4.2708,
+      "step": 296960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5129582543602015e-05,
+      "loss": 4.2849,
+      "step": 297472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5121196596091495e-05,
+      "loss": 4.2854,
+      "step": 297984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5112810648580975e-05,
+      "loss": 4.2755,
+      "step": 298496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5104424701070455e-05,
+      "loss": 4.2737,
+      "step": 299008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5096038753559935e-05,
+      "loss": 4.2823,
+      "step": 299520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5087652806049415e-05,
+      "loss": 4.2803,
+      "step": 300032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5079266858538895e-05,
+      "loss": 4.283,
+      "step": 300544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5070897289832104e-05,
+      "loss": 4.2661,
+      "step": 301056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5062511342321584e-05,
+      "loss": 4.2777,
+      "step": 301568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5054125394811064e-05,
+      "loss": 4.2758,
+      "step": 302080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5045739447300544e-05,
+      "loss": 4.2717,
+      "step": 302592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5037353499790024e-05,
+      "loss": 4.282,
+      "step": 303104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5028967552279503e-05,
+      "loss": 4.2672,
+      "step": 303616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5020581604768983e-05,
+      "loss": 4.2752,
+      "step": 304128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.501221203606219e-05,
+      "loss": 4.2728,
+      "step": 304640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.500382608855168e-05,
+      "loss": 4.2771,
+      "step": 305152
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.307571887969971,
+      "eval_runtime": 285.5289,
+      "eval_samples_per_second": 1336.436,
+      "eval_steps_per_second": 41.765,
+      "step": 305280
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.499544014104116e-05,
+      "loss": 4.2721,
+      "step": 305664
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.498705419353064e-05,
+      "loss": 4.271,
+      "step": 306176
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.497866824602012e-05,
+      "loss": 4.2783,
+      "step": 306688
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.49702822985096e-05,
+      "loss": 4.271,
+      "step": 307200
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.496189635099908e-05,
+      "loss": 4.2781,
+      "step": 307712
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.495351040348856e-05,
+      "loss": 4.2595,
+      "step": 308224
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.49451572135855e-05,
+      "loss": 4.2681,
+      "step": 308736
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.493677126607498e-05,
+      "loss": 4.2556,
+      "step": 309248
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.492838531856446e-05,
+      "loss": 4.2749,
+      "step": 309760
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.491999937105394e-05,
+      "loss": 4.2667,
+      "step": 310272
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4911629802347146e-05,
+      "loss": 4.2703,
+      "step": 310784
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.490324385483663e-05,
+      "loss": 4.2717,
+      "step": 311296
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.489487428612984e-05,
+      "loss": 4.2637,
+      "step": 311808
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.488648833861932e-05,
+      "loss": 4.2575,
+      "step": 312320
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.48781023911088e-05,
+      "loss": 4.2606,
+      "step": 312832
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.486971644359828e-05,
+      "loss": 4.2495,
+      "step": 313344
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.486133049608776e-05,
+      "loss": 4.2689,
+      "step": 313856
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.485294454857724e-05,
+      "loss": 4.2547,
+      "step": 314368
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.484455860106672e-05,
+      "loss": 4.2553,
+      "step": 314880
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.48361726535562e-05,
+      "loss": 4.285,
+      "step": 315392
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.482778670604568e-05,
+      "loss": 4.2577,
+      "step": 315904
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.481940075853516e-05,
+      "loss": 4.2631,
+      "step": 316416
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.481101481102464e-05,
+      "loss": 4.2649,
+      "step": 316928
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4802628863514115e-05,
+      "loss": 4.2625,
+      "step": 317440
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.479425929480733e-05,
+      "loss": 4.2596,
+      "step": 317952
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.478587334729681e-05,
+      "loss": 4.2488,
+      "step": 318464
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4777503778590026e-05,
+      "loss": 4.2644,
+      "step": 318976
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4769117831079506e-05,
+      "loss": 4.2421,
+      "step": 319488
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4760731883568986e-05,
+      "loss": 4.2533,
+      "step": 320000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4752345936058466e-05,
+      "loss": 4.2482,
+      "step": 320512
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.474395998854794e-05,
+      "loss": 4.2517,
+      "step": 321024
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.473557404103742e-05,
+      "loss": 4.2633,
+      "step": 321536
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.47271880935269e-05,
+      "loss": 4.2568,
+      "step": 322048
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.471880214601638e-05,
+      "loss": 4.2507,
+      "step": 322560
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4710448956113324e-05,
+      "loss": 4.259,
+      "step": 323072
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4702063008602804e-05,
+      "loss": 4.2635,
+      "step": 323584
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4693677061092284e-05,
+      "loss": 4.2474,
+      "step": 324096
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4685291113581764e-05,
+      "loss": 4.2559,
+      "step": 324608
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4676905166071244e-05,
+      "loss": 4.242,
+      "step": 325120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4668519218560724e-05,
+      "loss": 4.2579,
+      "step": 325632
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4660133271050204e-05,
+      "loss": 4.2372,
+      "step": 326144
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.465176370234341e-05,
+      "loss": 4.2645,
+      "step": 326656
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.464337775483289e-05,
+      "loss": 4.2527,
+      "step": 327168
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.463499180732237e-05,
+      "loss": 4.2499,
+      "step": 327680
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.462662223861559e-05,
+      "loss": 4.255,
+      "step": 328192
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.461823629110506e-05,
+      "loss": 4.2459,
+      "step": 328704
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.460985034359454e-05,
+      "loss": 4.2557,
+      "step": 329216
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.460146439608402e-05,
+      "loss": 4.2425,
+      "step": 329728
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.45930784485735e-05,
+      "loss": 4.2375,
+      "step": 330240
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.458469250106299e-05,
+      "loss": 4.2568,
+      "step": 330752
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.457630655355247e-05,
+      "loss": 4.2491,
+      "step": 331264
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.456792060604195e-05,
+      "loss": 4.2478,
+      "step": 331776
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.455953465853143e-05,
+      "loss": 4.2406,
+      "step": 332288
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.455114871102091e-05,
+      "loss": 4.2448,
+      "step": 332800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.454276276351039e-05,
+      "loss": 4.2385,
+      "step": 333312
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.45343931948036e-05,
+      "loss": 4.2516,
+      "step": 333824
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.452600724729308e-05,
+      "loss": 4.2408,
+      "step": 334336
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.451762129978256e-05,
+      "loss": 4.2545,
+      "step": 334848
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.450923535227204e-05,
+      "loss": 4.2477,
+      "step": 335360
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.450086578356525e-05,
+      "loss": 4.2369,
+      "step": 335872
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4492479836054727e-05,
+      "loss": 4.2375,
+      "step": 336384
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4484093888544207e-05,
+      "loss": 4.2456,
+      "step": 336896
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4475707941033686e-05,
+      "loss": 4.2282,
+      "step": 337408
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.446732199352317e-05,
+      "loss": 4.2377,
+      "step": 337920
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.445893604601265e-05,
+      "loss": 4.2414,
+      "step": 338432
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4450550098502126e-05,
+      "loss": 4.2447,
+      "step": 338944
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4442164150991606e-05,
+      "loss": 4.2218,
+      "step": 339456
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4433778203481086e-05,
+      "loss": 4.2311,
+      "step": 339968
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.44254086347743e-05,
+      "loss": 4.2239,
+      "step": 340480
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4417022687263775e-05,
+      "loss": 4.2496,
+      "step": 340992
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4408636739753255e-05,
+      "loss": 4.239,
+      "step": 341504
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4400250792242735e-05,
+      "loss": 4.2392,
+      "step": 342016
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4391864844732215e-05,
+      "loss": 4.239,
+      "step": 342528
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4383495276025424e-05,
+      "loss": 4.236,
+      "step": 343040
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.437510932851491e-05,
+      "loss": 4.254,
+      "step": 343552
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.436673975980813e-05,
+      "loss": 4.222,
+      "step": 344064
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.43583538122976e-05,
+      "loss": 4.2408,
+      "step": 344576
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.434996786478708e-05,
+      "loss": 4.2367,
+      "step": 345088
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.434158191727656e-05,
+      "loss": 4.2338,
+      "step": 345600
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.433319596976604e-05,
+      "loss": 4.2366,
+      "step": 346112
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.432481002225552e-05,
+      "loss": 4.233,
+      "step": 346624
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4316424074745e-05,
+      "loss": 4.2377,
+      "step": 347136
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.430803812723448e-05,
+      "loss": 4.2378,
+      "step": 347648
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.429965217972396e-05,
+      "loss": 4.2238,
+      "step": 348160
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.429128261101717e-05,
+      "loss": 4.2371,
+      "step": 348672
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.428289666350665e-05,
+      "loss": 4.2245,
+      "step": 349184
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4274527094799865e-05,
+      "loss": 4.2382,
+      "step": 349696
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4266141147289345e-05,
+      "loss": 4.2311,
+      "step": 350208
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4257755199778825e-05,
+      "loss": 4.219,
+      "step": 350720
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4249369252268305e-05,
+      "loss": 4.2317,
+      "step": 351232
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4240983304757784e-05,
+      "loss": 4.225,
+      "step": 351744
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4232613736050994e-05,
+      "loss": 4.2371,
+      "step": 352256
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4224227788540474e-05,
+      "loss": 4.2161,
+      "step": 352768
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4215841841029954e-05,
+      "loss": 4.2303,
+      "step": 353280
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4207455893519433e-05,
+      "loss": 4.2419,
+      "step": 353792
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4199069946008913e-05,
+      "loss": 4.2222,
+      "step": 354304
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.419068399849839e-05,
+      "loss": 4.2121,
+      "step": 354816
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.418229805098787e-05,
+      "loss": 4.2228,
+      "step": 355328
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.417391210347735e-05,
+      "loss": 4.2275,
+      "step": 355840
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.416554253477056e-05,
+      "loss": 4.233,
+      "step": 356352
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.415715658726004e-05,
+      "loss": 4.229,
+      "step": 356864
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.414878701855326e-05,
+      "loss": 4.22,
+      "step": 357376
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.414040107104274e-05,
+      "loss": 4.2172,
+      "step": 357888
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.413203150233595e-05,
+      "loss": 4.2304,
+      "step": 358400
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.412364555482543e-05,
+      "loss": 4.2063,
+      "step": 358912
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.411525960731491e-05,
+      "loss": 4.2199,
+      "step": 359424
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.410687365980439e-05,
+      "loss": 4.2218,
+      "step": 359936
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.409848771229387e-05,
+      "loss": 4.2251,
+      "step": 360448
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.409010176478335e-05,
+      "loss": 4.2031,
+      "step": 360960
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.408171581727283e-05,
+      "loss": 4.2224,
+      "step": 361472
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.407332986976231e-05,
+      "loss": 4.2143,
+      "step": 361984
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.406494392225179e-05,
+      "loss": 4.2156,
+      "step": 362496
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.405655797474127e-05,
+      "loss": 4.2174,
+      "step": 363008
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.404817202723075e-05,
+      "loss": 4.2078,
+      "step": 363520
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.403978607972023e-05,
+      "loss": 4.2221,
+      "step": 364032
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.403143288981717e-05,
+      "loss": 4.2232,
+      "step": 364544
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.402304694230665e-05,
+      "loss": 4.2143,
+      "step": 365056
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.401466099479613e-05,
+      "loss": 4.2223,
+      "step": 365568
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.400627504728561e-05,
+      "loss": 4.2112,
+      "step": 366080
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3997889099775085e-05,
+      "loss": 4.2234,
+      "step": 366592
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3989503152264565e-05,
+      "loss": 4.2056,
+      "step": 367104
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.398113358355778e-05,
+      "loss": 4.2272,
+      "step": 367616
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3972747636047254e-05,
+      "loss": 4.2065,
+      "step": 368128
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3964361688536734e-05,
+      "loss": 4.2158,
+      "step": 368640
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.395597574102622e-05,
+      "loss": 4.2094,
+      "step": 369152
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.39475897935157e-05,
+      "loss": 4.2212,
+      "step": 369664
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.393920384600518e-05,
+      "loss": 4.2113,
+      "step": 370176
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.393081789849466e-05,
+      "loss": 4.2099,
+      "step": 370688
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.392244832978787e-05,
+      "loss": 4.1965,
+      "step": 371200
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.391406238227735e-05,
+      "loss": 4.2259,
+      "step": 371712
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.390567643476683e-05,
+      "loss": 4.2132,
+      "step": 372224
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.389729048725631e-05,
+      "loss": 4.2164,
+      "step": 372736
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.388890453974579e-05,
+      "loss": 4.204,
+      "step": 373248
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.388051859223527e-05,
+      "loss": 4.219,
+      "step": 373760
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.387213264472475e-05,
+      "loss": 4.222,
+      "step": 374272
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.386374669721423e-05,
+      "loss": 4.2085,
+      "step": 374784
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.385537712850744e-05,
+      "loss": 4.207,
+      "step": 375296
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.384699118099692e-05,
+      "loss": 4.2166,
+      "step": 375808
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3838605233486405e-05,
+      "loss": 4.2164,
+      "step": 376320
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3830219285975885e-05,
+      "loss": 4.2181,
+      "step": 376832
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3821849717269094e-05,
+      "loss": 4.1979,
+      "step": 377344
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3813463769758574e-05,
+      "loss": 4.2114,
+      "step": 377856
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3805077822248054e-05,
+      "loss": 4.207,
+      "step": 378368
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3796691874737534e-05,
+      "loss": 4.2098,
+      "step": 378880
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.378832230603074e-05,
+      "loss": 4.216,
+      "step": 379392
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.377993635852022e-05,
+      "loss": 4.1985,
+      "step": 379904
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.37715504110097e-05,
+      "loss": 4.2167,
+      "step": 380416
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.376316446349918e-05,
+      "loss": 4.206,
+      "step": 380928
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.375477851598866e-05,
+      "loss": 4.2173,
+      "step": 381440
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.251643180847168,
+      "eval_runtime": 279.2584,
+      "eval_samples_per_second": 1366.444,
+      "eval_steps_per_second": 42.702,
+      "step": 381600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.374639256847814e-05,
+      "loss": 4.2005,
+      "step": 381952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.373800662096762e-05,
+      "loss": 4.2105,
+      "step": 382464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.37296206734571e-05,
+      "loss": 4.2118,
+      "step": 382976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.372123472594658e-05,
+      "loss": 4.2063,
+      "step": 383488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.371284877843606e-05,
+      "loss": 4.2123,
+      "step": 384000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.370446283092554e-05,
+      "loss": 4.2002,
+      "step": 384512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.369607688341502e-05,
+      "loss": 4.2033,
+      "step": 385024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.36876909359045e-05,
+      "loss": 4.1939,
+      "step": 385536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.367930498839398e-05,
+      "loss": 4.2104,
+      "step": 386048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.367091904088346e-05,
+      "loss": 4.2075,
+      "step": 386560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.366253309337294e-05,
+      "loss": 4.2057,
+      "step": 387072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.365414714586242e-05,
+      "loss": 4.2094,
+      "step": 387584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.364577757715563e-05,
+      "loss": 4.1978,
+      "step": 388096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.363739162964511e-05,
+      "loss": 4.1964,
+      "step": 388608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.362900568213459e-05,
+      "loss": 4.1986,
+      "step": 389120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.362061973462407e-05,
+      "loss": 4.1871,
+      "step": 389632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.361223378711355e-05,
+      "loss": 4.2023,
+      "step": 390144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.360386421840677e-05,
+      "loss": 4.1914,
+      "step": 390656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.359547827089625e-05,
+      "loss": 4.1942,
+      "step": 391168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.358709232338573e-05,
+      "loss": 4.2279,
+      "step": 391680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.357870637587521e-05,
+      "loss": 4.1941,
+      "step": 392192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.357032042836469e-05,
+      "loss": 4.2019,
+      "step": 392704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.356193448085416e-05,
+      "loss": 4.2025,
+      "step": 393216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.355354853334364e-05,
+      "loss": 4.202,
+      "step": 393728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.354516258583312e-05,
+      "loss": 4.194,
+      "step": 394240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.35367766383226e-05,
+      "loss": 4.1912,
+      "step": 394752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.352840706961581e-05,
+      "loss": 4.1999,
+      "step": 395264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.352002112210529e-05,
+      "loss": 4.183,
+      "step": 395776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.3511635174594776e-05,
+      "loss": 4.1925,
+      "step": 396288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.3503249227084256e-05,
+      "loss": 4.187,
+      "step": 396800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3494863279573736e-05,
+      "loss": 4.1935,
+      "step": 397312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3486477332063216e-05,
+      "loss": 4.1997,
+      "step": 397824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3478091384552696e-05,
+      "loss": 4.1995,
+      "step": 398336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3469705437042175e-05,
+      "loss": 4.1903,
+      "step": 398848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3461352247139114e-05,
+      "loss": 4.1969,
+      "step": 399360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3452966299628594e-05,
+      "loss": 4.2002,
+      "step": 399872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3444580352118074e-05,
+      "loss": 4.1864,
+      "step": 400384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3436194404607554e-05,
+      "loss": 4.1984,
+      "step": 400896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.342782483590076e-05,
+      "loss": 4.1776,
+      "step": 401408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.341943888839024e-05,
+      "loss": 4.2019,
+      "step": 401920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.341105294087973e-05,
+      "loss": 4.1743,
+      "step": 402432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.340266699336921e-05,
+      "loss": 4.208,
+      "step": 402944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.339428104585869e-05,
+      "loss": 4.1936,
+      "step": 403456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.338589509834817e-05,
+      "loss": 4.1923,
+      "step": 403968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.337750915083765e-05,
+      "loss": 4.1951,
+      "step": 404480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.336912320332713e-05,
+      "loss": 4.1892,
+      "step": 404992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.336075363462034e-05,
+      "loss": 4.1937,
+      "step": 405504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.335236768710982e-05,
+      "loss": 4.181,
+      "step": 406016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.33439817395993e-05,
+      "loss": 4.1736,
+      "step": 406528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.333559579208878e-05,
+      "loss": 4.2034,
+      "step": 407040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.332720984457826e-05,
+      "loss": 4.1898,
+      "step": 407552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.331884027587147e-05,
+      "loss": 4.1966,
+      "step": 408064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.331045432836095e-05,
+      "loss": 4.182,
+      "step": 408576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.330206838085043e-05,
+      "loss": 4.1873,
+      "step": 409088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3293682433339914e-05,
+      "loss": 4.1812,
+      "step": 409600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3285296485829394e-05,
+      "loss": 4.192,
+      "step": 410112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3276910538318874e-05,
+      "loss": 4.1828,
+      "step": 410624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.326852459080835e-05,
+      "loss": 4.1978,
+      "step": 411136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.326013864329783e-05,
+      "loss": 4.1953,
+      "step": 411648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.325176907459104e-05,
+      "loss": 4.1773,
+      "step": 412160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.324338312708052e-05,
+      "loss": 4.18,
+      "step": 412672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3234997179569996e-05,
+      "loss": 4.191,
+      "step": 413184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3226611232059476e-05,
+      "loss": 4.1737,
+      "step": 413696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.321824166335269e-05,
+      "loss": 4.1785,
+      "step": 414208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3209855715842165e-05,
+      "loss": 4.1839,
+      "step": 414720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.320146976833165e-05,
+      "loss": 4.1849,
+      "step": 415232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.319308382082113e-05,
+      "loss": 4.1677,
+      "step": 415744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.318469787331061e-05,
+      "loss": 4.1751,
+      "step": 416256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.317631192580009e-05,
+      "loss": 4.1669,
+      "step": 416768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.316792597828957e-05,
+      "loss": 4.1924,
+      "step": 417280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.315955640958278e-05,
+      "loss": 4.1834,
+      "step": 417792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.315117046207226e-05,
+      "loss": 4.1817,
+      "step": 418304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.314278451456174e-05,
+      "loss": 4.1869,
+      "step": 418816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.313439856705122e-05,
+      "loss": 4.177,
+      "step": 419328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.312602899834443e-05,
+      "loss": 4.197,
+      "step": 419840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.311765942963764e-05,
+      "loss": 4.1677,
+      "step": 420352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.310927348212712e-05,
+      "loss": 4.1843,
+      "step": 420864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3100887534616605e-05,
+      "loss": 4.1807,
+      "step": 421376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3092501587106085e-05,
+      "loss": 4.1793,
+      "step": 421888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3084115639595565e-05,
+      "loss": 4.1823,
+      "step": 422400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3075729692085045e-05,
+      "loss": 4.1776,
+      "step": 422912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3067343744574525e-05,
+      "loss": 4.1795,
+      "step": 423424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3058957797064005e-05,
+      "loss": 4.1865,
+      "step": 423936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3050588228357214e-05,
+      "loss": 4.1681,
+      "step": 424448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3042202280846694e-05,
+      "loss": 4.1839,
+      "step": 424960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3033816333336174e-05,
+      "loss": 4.172,
+      "step": 425472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3025430385825654e-05,
+      "loss": 4.1837,
+      "step": 425984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.301706081711886e-05,
+      "loss": 4.1717,
+      "step": 426496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.300867486960834e-05,
+      "loss": 4.1686,
+      "step": 427008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.300028892209782e-05,
+      "loss": 4.1756,
+      "step": 427520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.29919029745873e-05,
+      "loss": 4.1747,
+      "step": 428032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.298353340588052e-05,
+      "loss": 4.1775,
+      "step": 428544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.297514745837e-05,
+      "loss": 4.1639,
+      "step": 429056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.296676151085948e-05,
+      "loss": 4.1771,
+      "step": 429568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.295837556334896e-05,
+      "loss": 4.186,
+      "step": 430080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.295000599464217e-05,
+      "loss": 4.173,
+      "step": 430592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.294162004713165e-05,
+      "loss": 4.1559,
+      "step": 431104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.293325047842486e-05,
+      "loss": 4.1709,
+      "step": 431616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.292486453091434e-05,
+      "loss": 4.1735,
+      "step": 432128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.291647858340382e-05,
+      "loss": 4.1771,
+      "step": 432640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.29080926358933e-05,
+      "loss": 4.1784,
+      "step": 433152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.289970668838278e-05,
+      "loss": 4.1684,
+      "step": 433664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.289132074087226e-05,
+      "loss": 4.1642,
+      "step": 434176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2882934793361737e-05,
+      "loss": 4.1768,
+      "step": 434688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.287454884585122e-05,
+      "loss": 4.1544,
+      "step": 435200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.286617927714443e-05,
+      "loss": 4.1667,
+      "step": 435712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.285780970843764e-05,
+      "loss": 4.1664,
+      "step": 436224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.284942376092712e-05,
+      "loss": 4.1742,
+      "step": 436736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.28410378134166e-05,
+      "loss": 4.152,
+      "step": 437248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.283265186590608e-05,
+      "loss": 4.1657,
+      "step": 437760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.282428229719929e-05,
+      "loss": 4.167,
+      "step": 438272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.281589634968877e-05,
+      "loss": 4.1619,
+      "step": 438784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.280751040217825e-05,
+      "loss": 4.1683,
+      "step": 439296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.279912445466773e-05,
+      "loss": 4.1549,
+      "step": 439808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.279073850715721e-05,
+      "loss": 4.1723,
+      "step": 440320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2782368938450426e-05,
+      "loss": 4.1711,
+      "step": 440832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2773982990939906e-05,
+      "loss": 4.1599,
+      "step": 441344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2765597043429386e-05,
+      "loss": 4.173,
+      "step": 441856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2757211095918866e-05,
+      "loss": 4.1582,
+      "step": 442368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2748825148408346e-05,
+      "loss": 4.1752,
+      "step": 442880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2740439200897826e-05,
+      "loss": 4.1571,
+      "step": 443392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.27320532533873e-05,
+      "loss": 4.1707,
+      "step": 443904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.272366730587678e-05,
+      "loss": 4.1556,
+      "step": 444416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2715297737169995e-05,
+      "loss": 4.1648,
+      "step": 444928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2706911789659475e-05,
+      "loss": 4.1618,
+      "step": 445440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.269852584214895e-05,
+      "loss": 4.1664,
+      "step": 445952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.269013989463843e-05,
+      "loss": 4.1588,
+      "step": 446464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2681770325931644e-05,
+      "loss": 4.1659,
+      "step": 446976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.267338437842113e-05,
+      "loss": 4.1435,
+      "step": 447488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2664998430910604e-05,
+      "loss": 4.1792,
+      "step": 448000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2656612483400084e-05,
+      "loss": 4.1607,
+      "step": 448512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2648226535889564e-05,
+      "loss": 4.1653,
+      "step": 449024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.263985696718277e-05,
+      "loss": 4.1529,
+      "step": 449536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.263148739847599e-05,
+      "loss": 4.1697,
+      "step": 450048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.262310145096547e-05,
+      "loss": 4.1703,
+      "step": 450560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.261471550345495e-05,
+      "loss": 4.1575,
+      "step": 451072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.260632955594442e-05,
+      "loss": 4.1611,
+      "step": 451584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.25979436084339e-05,
+      "loss": 4.1659,
+      "step": 452096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.258955766092338e-05,
+      "loss": 4.1647,
+      "step": 452608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.258117171341287e-05,
+      "loss": 4.1745,
+      "step": 453120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.257278576590235e-05,
+      "loss": 4.1454,
+      "step": 453632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.256439981839183e-05,
+      "loss": 4.1668,
+      "step": 454144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.255601387088131e-05,
+      "loss": 4.1538,
+      "step": 454656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.254762792337079e-05,
+      "loss": 4.1636,
+      "step": 455168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2539258354664e-05,
+      "loss": 4.1627,
+      "step": 455680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.253087240715348e-05,
+      "loss": 4.1458,
+      "step": 456192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.252248645964296e-05,
+      "loss": 4.1694,
+      "step": 456704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.251410051213244e-05,
+      "loss": 4.1598,
+      "step": 457216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.250571456462192e-05,
+      "loss": 4.1646,
+      "step": 457728
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.2104573249816895,
+      "eval_runtime": 295.6006,
+      "eval_samples_per_second": 1290.9,
+      "eval_steps_per_second": 40.342,
+      "step": 457920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2497344995915126e-05,
+      "loss": 4.1566,
+      "step": 458240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2488959048404606e-05,
+      "loss": 4.1578,
+      "step": 458752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2480573100894086e-05,
+      "loss": 4.1658,
+      "step": 459264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2472187153383566e-05,
+      "loss": 4.1593,
+      "step": 459776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2463801205873046e-05,
+      "loss": 4.1627,
+      "step": 460288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.245541525836253e-05,
+      "loss": 4.1554,
+      "step": 460800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.244702931085201e-05,
+      "loss": 4.1487,
+      "step": 461312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.243864336334149e-05,
+      "loss": 4.151,
+      "step": 461824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2430257415830966e-05,
+      "loss": 4.1586,
+      "step": 462336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2421871468320446e-05,
+      "loss": 4.163,
+      "step": 462848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2413485520809926e-05,
+      "loss": 4.1569,
+      "step": 463360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2405099573299406e-05,
+      "loss": 4.1615,
+      "step": 463872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2396730004592615e-05,
+      "loss": 4.1483,
+      "step": 464384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2388344057082095e-05,
+      "loss": 4.1516,
+      "step": 464896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2379958109571575e-05,
+      "loss": 4.152,
+      "step": 465408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2371572162061055e-05,
+      "loss": 4.1386,
+      "step": 465920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2363186214550535e-05,
+      "loss": 4.1551,
+      "step": 466432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2354800267040015e-05,
+      "loss": 4.1407,
+      "step": 466944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.23464143195295e-05,
+      "loss": 4.1467,
+      "step": 467456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.233802837201898e-05,
+      "loss": 4.1821,
+      "step": 467968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.232964242450846e-05,
+      "loss": 4.148,
+      "step": 468480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.232125647699794e-05,
+      "loss": 4.1514,
+      "step": 468992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.231287052948742e-05,
+      "loss": 4.1546,
+      "step": 469504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.23044845819769e-05,
+      "loss": 4.1601,
+      "step": 470016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.229611501327011e-05,
+      "loss": 4.1393,
+      "step": 470528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.228772906575959e-05,
+      "loss": 4.1473,
+      "step": 471040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.227934311824907e-05,
+      "loss": 4.1509,
+      "step": 471552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.227095717073855e-05,
+      "loss": 4.1415,
+      "step": 472064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.226257122322802e-05,
+      "loss": 4.1454,
+      "step": 472576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.22541852757175e-05,
+      "loss": 4.1405,
+      "step": 473088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.224581570701072e-05,
+      "loss": 4.1479,
+      "step": 473600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.22374297595002e-05,
+      "loss": 4.1531,
+      "step": 474112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.222904381198968e-05,
+      "loss": 4.1498,
+      "step": 474624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.222065786447916e-05,
+      "loss": 4.1448,
+      "step": 475136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2212288295772375e-05,
+      "loss": 4.1499,
+      "step": 475648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.220390234826185e-05,
+      "loss": 4.1546,
+      "step": 476160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2195532779555064e-05,
+      "loss": 4.1404,
+      "step": 476672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2187146832044544e-05,
+      "loss": 4.157,
+      "step": 477184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2178760884534024e-05,
+      "loss": 4.1316,
+      "step": 477696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.21703749370235e-05,
+      "loss": 4.152,
+      "step": 478208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.216198898951298e-05,
+      "loss": 4.1307,
+      "step": 478720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.215360304200246e-05,
+      "loss": 4.1613,
+      "step": 479232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.214521709449194e-05,
+      "loss": 4.1459,
+      "step": 479744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2136831146981424e-05,
+      "loss": 4.1499,
+      "step": 480256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.212847795707837e-05,
+      "loss": 4.1484,
+      "step": 480768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.212010838837158e-05,
+      "loss": 4.1457,
+      "step": 481280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.211172244086106e-05,
+      "loss": 4.1502,
+      "step": 481792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.210333649335054e-05,
+      "loss": 4.1396,
+      "step": 482304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.209495054584002e-05,
+      "loss": 4.1252,
+      "step": 482816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.20865645983295e-05,
+      "loss": 4.1597,
+      "step": 483328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.207817865081897e-05,
+      "loss": 4.1417,
+      "step": 483840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.206979270330845e-05,
+      "loss": 4.1538,
+      "step": 484352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.206140675579793e-05,
+      "loss": 4.1339,
+      "step": 484864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.205302080828741e-05,
+      "loss": 4.1411,
+      "step": 485376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.204463486077689e-05,
+      "loss": 4.1339,
+      "step": 485888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.203624891326638e-05,
+      "loss": 4.1492,
+      "step": 486400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.202786296575586e-05,
+      "loss": 4.1415,
+      "step": 486912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.201947701824534e-05,
+      "loss": 4.1502,
+      "step": 487424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2011107449538546e-05,
+      "loss": 4.1492,
+      "step": 487936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2002737880831755e-05,
+      "loss": 4.1326,
+      "step": 488448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1994351933321235e-05,
+      "loss": 4.134,
+      "step": 488960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1985965985810715e-05,
+      "loss": 4.1516,
+      "step": 489472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1977580038300195e-05,
+      "loss": 4.127,
+      "step": 489984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1969194090789675e-05,
+      "loss": 4.1338,
+      "step": 490496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1960808143279155e-05,
+      "loss": 4.1392,
+      "step": 491008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1952422195768635e-05,
+      "loss": 4.1474,
+      "step": 491520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1944036248258115e-05,
+      "loss": 4.1235,
+      "step": 492032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1935666679551324e-05,
+      "loss": 4.1294,
+      "step": 492544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.192728073204081e-05,
+      "loss": 4.1277,
+      "step": 493056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.191889478453029e-05,
+      "loss": 4.1384,
+      "step": 493568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.191050883701977e-05,
+      "loss": 4.1437,
+      "step": 494080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.190213926831298e-05,
+      "loss": 4.1395,
+      "step": 494592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.189375332080246e-05,
+      "loss": 4.1403,
+      "step": 495104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.188536737329194e-05,
+      "loss": 4.1331,
+      "step": 495616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.187699780458515e-05,
+      "loss": 4.1567,
+      "step": 496128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.186861185707463e-05,
+      "loss": 4.1253,
+      "step": 496640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.186022590956411e-05,
+      "loss": 4.1394,
+      "step": 497152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.185183996205359e-05,
+      "loss": 4.1377,
+      "step": 497664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.18434703933468e-05,
+      "loss": 4.1353,
+      "step": 498176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.183508444583628e-05,
+      "loss": 4.1423,
+      "step": 498688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1826698498325765e-05,
+      "loss": 4.1367,
+      "step": 499200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1818312550815245e-05,
+      "loss": 4.1324,
+      "step": 499712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1809926603304725e-05,
+      "loss": 4.1459,
+      "step": 500224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1801540655794204e-05,
+      "loss": 4.1274,
+      "step": 500736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1793154708283684e-05,
+      "loss": 4.1377,
+      "step": 501248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.178476876077316e-05,
+      "loss": 4.1313,
+      "step": 501760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.17764155708701e-05,
+      "loss": 4.1407,
+      "step": 502272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.176802962335958e-05,
+      "loss": 4.13,
+      "step": 502784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.175964367584906e-05,
+      "loss": 4.1269,
+      "step": 503296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.175125772833854e-05,
+      "loss": 4.132,
+      "step": 503808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.174287178082802e-05,
+      "loss": 4.135,
+      "step": 504320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.17344858333175e-05,
+      "loss": 4.1362,
+      "step": 504832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.172609988580698e-05,
+      "loss": 4.1214,
+      "step": 505344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.17177303171002e-05,
+      "loss": 4.1369,
+      "step": 505856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.170934436958968e-05,
+      "loss": 4.1425,
+      "step": 506368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.170095842207916e-05,
+      "loss": 4.1333,
+      "step": 506880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.169257247456863e-05,
+      "loss": 4.1144,
+      "step": 507392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.168418652705811e-05,
+      "loss": 4.1283,
+      "step": 507904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.167580057954759e-05,
+      "loss": 4.1273,
+      "step": 508416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.166741463203707e-05,
+      "loss": 4.139,
+      "step": 508928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.165902868452655e-05,
+      "loss": 4.1394,
+      "step": 509440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.165064273701603e-05,
+      "loss": 4.1254,
+      "step": 509952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.164227316830924e-05,
+      "loss": 4.1238,
+      "step": 510464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.163388722079872e-05,
+      "loss": 4.1348,
+      "step": 510976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1625517652091936e-05,
+      "loss": 4.1142,
+      "step": 511488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1617131704581416e-05,
+      "loss": 4.1283,
+      "step": 512000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1608745757070896e-05,
+      "loss": 4.121,
+      "step": 512512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1600359809560376e-05,
+      "loss": 4.1383,
+      "step": 513024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1591973862049856e-05,
+      "loss": 4.1124,
+      "step": 513536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1583587914539336e-05,
+      "loss": 4.1207,
+      "step": 514048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1575201967028816e-05,
+      "loss": 4.1223,
+      "step": 514560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1566816019518296e-05,
+      "loss": 4.1276,
+      "step": 515072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1558462829615234e-05,
+      "loss": 4.1224,
+      "step": 515584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1550076882104714e-05,
+      "loss": 4.1153,
+      "step": 516096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1541690934594194e-05,
+      "loss": 4.1289,
+      "step": 516608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1533304987083674e-05,
+      "loss": 4.1318,
+      "step": 517120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1524919039573154e-05,
+      "loss": 4.1196,
+      "step": 517632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.151654947086637e-05,
+      "loss": 4.1352,
+      "step": 518144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.150816352335585e-05,
+      "loss": 4.1165,
+      "step": 518656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.149977757584533e-05,
+      "loss": 4.1272,
+      "step": 519168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.149139162833481e-05,
+      "loss": 4.1224,
+      "step": 519680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.148302205962802e-05,
+      "loss": 4.127,
+      "step": 520192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.14746361121175e-05,
+      "loss": 4.1172,
+      "step": 520704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.146625016460698e-05,
+      "loss": 4.124,
+      "step": 521216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.145786421709646e-05,
+      "loss": 4.1267,
+      "step": 521728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.144947826958594e-05,
+      "loss": 4.1223,
+      "step": 522240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.144109232207542e-05,
+      "loss": 4.1171,
+      "step": 522752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.14327063745649e-05,
+      "loss": 4.1256,
+      "step": 523264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.142432042705438e-05,
+      "loss": 4.1075,
+      "step": 523776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.141593447954386e-05,
+      "loss": 4.1364,
+      "step": 524288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.140754853203334e-05,
+      "loss": 4.1215,
+      "step": 524800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1399178963326554e-05,
+      "loss": 4.1249,
+      "step": 525312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1390793015816034e-05,
+      "loss": 4.115,
+      "step": 525824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1382407068305514e-05,
+      "loss": 4.1359,
+      "step": 526336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1374021120794994e-05,
+      "loss": 4.1253,
+      "step": 526848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.13656515520882e-05,
+      "loss": 4.1189,
+      "step": 527360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.135726560457768e-05,
+      "loss": 4.1203,
+      "step": 527872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.134887965706716e-05,
+      "loss": 4.1251,
+      "step": 528384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.134049370955664e-05,
+      "loss": 4.1246,
+      "step": 528896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.133212414084985e-05,
+      "loss": 4.1353,
+      "step": 529408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.132375457214306e-05,
+      "loss": 4.1062,
+      "step": 529920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.131536862463254e-05,
+      "loss": 4.1247,
+      "step": 530432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.130698267712203e-05,
+      "loss": 4.1175,
+      "step": 530944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.129859672961151e-05,
+      "loss": 4.1271,
+      "step": 531456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.129022716090472e-05,
+      "loss": 4.1186,
+      "step": 531968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.12818412133942e-05,
+      "loss": 4.1111,
+      "step": 532480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.127345526588368e-05,
+      "loss": 4.1266,
+      "step": 532992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.126506931837316e-05,
+      "loss": 4.1211,
+      "step": 533504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.125668337086264e-05,
+      "loss": 4.1238,
+      "step": 534016
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.1807756423950195,
+      "eval_runtime": 292.2925,
+      "eval_samples_per_second": 1305.511,
+      "eval_steps_per_second": 40.798,
+      "step": 534240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1248313802155846e-05,
+      "loss": 4.1209,
+      "step": 534528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1239927854645326e-05,
+      "loss": 4.1163,
+      "step": 535040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1231541907134806e-05,
+      "loss": 4.1258,
+      "step": 535552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1223155959624286e-05,
+      "loss": 4.1244,
+      "step": 536064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1214770012113766e-05,
+      "loss": 4.1285,
+      "step": 536576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1206384064603246e-05,
+      "loss": 4.113,
+      "step": 537088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.119801449589646e-05,
+      "loss": 4.1117,
+      "step": 537600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.118962854838594e-05,
+      "loss": 4.1076,
+      "step": 538112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1181242600875415e-05,
+      "loss": 4.1223,
+      "step": 538624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1172856653364894e-05,
+      "loss": 4.1272,
+      "step": 539136
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1164470705854374e-05,
+      "loss": 4.1145,
+      "step": 539648
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.115610113714759e-05,
+      "loss": 4.1237,
+      "step": 540160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.11477315684408e-05,
+      "loss": 4.1132,
+      "step": 540672
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.113934562093028e-05,
+      "loss": 4.112,
+      "step": 541184
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.113095967341976e-05,
+      "loss": 4.1108,
+      "step": 541696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.112257372590923e-05,
+      "loss": 4.1071,
+      "step": 542208
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.111418777839872e-05,
+      "loss": 4.1131,
+      "step": 542720
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.11058018308882e-05,
+      "loss": 4.1066,
+      "step": 543232
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.109741588337768e-05,
+      "loss": 4.1066,
+      "step": 543744
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.108902993586716e-05,
+      "loss": 4.1397,
+      "step": 544256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.108064398835664e-05,
+      "loss": 4.1146,
+      "step": 544768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.107225804084612e-05,
+      "loss": 4.1122,
+      "step": 545280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.10638720933356e-05,
+      "loss": 4.1162,
+      "step": 545792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.105548614582508e-05,
+      "loss": 4.1197,
+      "step": 546304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.104711657711829e-05,
+      "loss": 4.1016,
+      "step": 546816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.103873062960777e-05,
+      "loss": 4.1115,
+      "step": 547328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.103036106090098e-05,
+      "loss": 4.1148,
+      "step": 547840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.102197511339046e-05,
+      "loss": 4.1079,
+      "step": 548352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.101358916587994e-05,
+      "loss": 4.1044,
+      "step": 548864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.100520321836942e-05,
+      "loss": 4.1036,
+      "step": 549376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0996817270858904e-05,
+      "loss": 4.1108,
+      "step": 549888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0988431323348384e-05,
+      "loss": 4.1184,
+      "step": 550400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0980045375837864e-05,
+      "loss": 4.1139,
+      "step": 550912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0971659428327344e-05,
+      "loss": 4.1067,
+      "step": 551424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.096330623842428e-05,
+      "loss": 4.1131,
+      "step": 551936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.095492029091376e-05,
+      "loss": 4.113,
+      "step": 552448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.094653434340324e-05,
+      "loss": 4.1086,
+      "step": 552960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.093814839589272e-05,
+      "loss": 4.1145,
+      "step": 553472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.09297624483822e-05,
+      "loss": 4.0967,
+      "step": 553984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.092137650087168e-05,
+      "loss": 4.1135,
+      "step": 554496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.091299055336116e-05,
+      "loss": 4.0969,
+      "step": 555008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.090460460585064e-05,
+      "loss": 4.1215,
+      "step": 555520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.089623503714386e-05,
+      "loss": 4.1121,
+      "step": 556032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.088784908963334e-05,
+      "loss": 4.1119,
+      "step": 556544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.087946314212282e-05,
+      "loss": 4.1152,
+      "step": 557056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.08710771946123e-05,
+      "loss": 4.1071,
+      "step": 557568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.086269124710178e-05,
+      "loss": 4.1147,
+      "step": 558080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0854321678394986e-05,
+      "loss": 4.1048,
+      "step": 558592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0845935730884466e-05,
+      "loss": 4.0852,
+      "step": 559104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0837549783373946e-05,
+      "loss": 4.1232,
+      "step": 559616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.082916383586342e-05,
+      "loss": 4.1055,
+      "step": 560128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0820794267156635e-05,
+      "loss": 4.1186,
+      "step": 560640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0812408319646115e-05,
+      "loss": 4.1016,
+      "step": 561152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0804022372135595e-05,
+      "loss": 4.1047,
+      "step": 561664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0795636424625075e-05,
+      "loss": 4.0932,
+      "step": 562176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.078726685591829e-05,
+      "loss": 4.1145,
+      "step": 562688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.077888090840777e-05,
+      "loss": 4.1072,
+      "step": 563200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.077051133970098e-05,
+      "loss": 4.1106,
+      "step": 563712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.076212539219046e-05,
+      "loss": 4.1197,
+      "step": 564224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.075373944467994e-05,
+      "loss": 4.0972,
+      "step": 564736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.074535349716942e-05,
+      "loss": 4.0981,
+      "step": 565248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.073696754965889e-05,
+      "loss": 4.1159,
+      "step": 565760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.072858160214837e-05,
+      "loss": 4.094,
+      "step": 566272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.072019565463785e-05,
+      "loss": 4.0964,
+      "step": 566784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.071180970712733e-05,
+      "loss": 4.104,
+      "step": 567296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.070344013842055e-05,
+      "loss": 4.1099,
+      "step": 567808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.069505419091003e-05,
+      "loss": 4.0913,
+      "step": 568320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.068666824339951e-05,
+      "loss": 4.0954,
+      "step": 568832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.067828229588899e-05,
+      "loss": 4.0892,
+      "step": 569344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.066989634837847e-05,
+      "loss": 4.105,
+      "step": 569856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.066152677967168e-05,
+      "loss": 4.1086,
+      "step": 570368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.065314083216116e-05,
+      "loss": 4.1053,
+      "step": 570880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.064475488465064e-05,
+      "loss": 4.1012,
+      "step": 571392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.063636893714012e-05,
+      "loss": 4.1023,
+      "step": 571904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.062799936843333e-05,
+      "loss": 4.121,
+      "step": 572416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.061961342092281e-05,
+      "loss": 4.09,
+      "step": 572928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0611227473412287e-05,
+      "loss": 4.1092,
+      "step": 573440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0602841525901767e-05,
+      "loss": 4.1033,
+      "step": 573952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.059447195719498e-05,
+      "loss": 4.0954,
+      "step": 574464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.058608600968446e-05,
+      "loss": 4.1095,
+      "step": 574976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.057770006217394e-05,
+      "loss": 4.1024,
+      "step": 575488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.056931411466342e-05,
+      "loss": 4.0995,
+      "step": 576000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.056094454595663e-05,
+      "loss": 4.1101,
+      "step": 576512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.055255859844611e-05,
+      "loss": 4.092,
+      "step": 577024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.054417265093559e-05,
+      "loss": 4.1016,
+      "step": 577536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.053578670342507e-05,
+      "loss": 4.0992,
+      "step": 578048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.052741713471828e-05,
+      "loss": 4.1077,
+      "step": 578560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.051904756601149e-05,
+      "loss": 4.0937,
+      "step": 579072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.051066161850097e-05,
+      "loss": 4.0971,
+      "step": 579584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.050227567099045e-05,
+      "loss": 4.0907,
+      "step": 580096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0493889723479936e-05,
+      "loss": 4.1023,
+      "step": 580608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0485503775969416e-05,
+      "loss": 4.1018,
+      "step": 581120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0477117828458896e-05,
+      "loss": 4.0855,
+      "step": 581632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0468731880948376e-05,
+      "loss": 4.1044,
+      "step": 582144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0460345933437856e-05,
+      "loss": 4.1048,
+      "step": 582656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0451959985927336e-05,
+      "loss": 4.0978,
+      "step": 583168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0443574038416816e-05,
+      "loss": 4.0804,
+      "step": 583680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0435188090906296e-05,
+      "loss": 4.0954,
+      "step": 584192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0426818522199505e-05,
+      "loss": 4.0921,
+      "step": 584704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0418432574688985e-05,
+      "loss": 4.1061,
+      "step": 585216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0410046627178465e-05,
+      "loss": 4.1037,
+      "step": 585728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0401660679667945e-05,
+      "loss": 4.0913,
+      "step": 586240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0393274732157425e-05,
+      "loss": 4.0934,
+      "step": 586752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0384888784646905e-05,
+      "loss": 4.0939,
+      "step": 587264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0376502837136385e-05,
+      "loss": 4.0875,
+      "step": 587776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0368116889625865e-05,
+      "loss": 4.0917,
+      "step": 588288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.035974732091908e-05,
+      "loss": 4.086,
+      "step": 588800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0351361373408554e-05,
+      "loss": 4.1052,
+      "step": 589312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0342975425898034e-05,
+      "loss": 4.0782,
+      "step": 589824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.033460585719125e-05,
+      "loss": 4.0872,
+      "step": 590336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.032621990968073e-05,
+      "loss": 4.0868,
+      "step": 590848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.03178339621702e-05,
+      "loss": 4.0966,
+      "step": 591360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.030944801465968e-05,
+      "loss": 4.0875,
+      "step": 591872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.030106206714916e-05,
+      "loss": 4.083,
+      "step": 592384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.029267611963864e-05,
+      "loss": 4.0964,
+      "step": 592896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.028430655093186e-05,
+      "loss": 4.0988,
+      "step": 593408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.027592060342134e-05,
+      "loss": 4.0836,
+      "step": 593920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.026753465591082e-05,
+      "loss": 4.0994,
+      "step": 594432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.02591487084003e-05,
+      "loss": 4.0872,
+      "step": 594944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.025076276088978e-05,
+      "loss": 4.0935,
+      "step": 595456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.024237681337926e-05,
+      "loss": 4.0887,
+      "step": 595968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.023400724467247e-05,
+      "loss": 4.0929,
+      "step": 596480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.022562129716195e-05,
+      "loss": 4.0858,
+      "step": 596992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.021723534965143e-05,
+      "loss": 4.094,
+      "step": 597504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.020884940214091e-05,
+      "loss": 4.0955,
+      "step": 598016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.020046345463039e-05,
+      "loss": 4.0909,
+      "step": 598528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.019207750711987e-05,
+      "loss": 4.0856,
+      "step": 599040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.018369155960935e-05,
+      "loss": 4.0919,
+      "step": 599552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.017530561209883e-05,
+      "loss": 4.0734,
+      "step": 600064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.016693604339204e-05,
+      "loss": 4.1065,
+      "step": 600576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.015855009588152e-05,
+      "loss": 4.0851,
+      "step": 601088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0150164148371e-05,
+      "loss": 4.0924,
+      "step": 601600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.014179457966421e-05,
+      "loss": 4.0807,
+      "step": 602112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.013340863215369e-05,
+      "loss": 4.1009,
+      "step": 602624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.012502268464317e-05,
+      "loss": 4.0942,
+      "step": 603136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.011663673713265e-05,
+      "loss": 4.0876,
+      "step": 603648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.010825078962213e-05,
+      "loss": 4.0915,
+      "step": 604160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.009986484211161e-05,
+      "loss": 4.0891,
+      "step": 604672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.009147889460109e-05,
+      "loss": 4.0912,
+      "step": 605184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0083092947090565e-05,
+      "loss": 4.108,
+      "step": 605696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0074706999580045e-05,
+      "loss": 4.0711,
+      "step": 606208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.006633743087326e-05,
+      "loss": 4.0985,
+      "step": 606720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.005795148336274e-05,
+      "loss": 4.0846,
+      "step": 607232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0049581914655956e-05,
+      "loss": 4.0948,
+      "step": 607744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0041195967145436e-05,
+      "loss": 4.0853,
+      "step": 608256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0032810019634916e-05,
+      "loss": 4.0854,
+      "step": 608768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.002442407212439e-05,
+      "loss": 4.0924,
+      "step": 609280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.001603812461387e-05,
+      "loss": 4.0904,
+      "step": 609792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.000765217710335e-05,
+      "loss": 4.0899,
+      "step": 610304
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.156196594238281,
+      "eval_runtime": 292.4597,
+      "eval_samples_per_second": 1304.764,
+      "eval_steps_per_second": 40.775,
+      "step": 610560
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.999926622959283e-05,
+      "loss": 4.0896,
+      "step": 610816
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.999088028208231e-05,
+      "loss": 4.086,
+      "step": 611328
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.998251071337552e-05,
+      "loss": 4.0926,
+      "step": 611840
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.9974124765865e-05,
+      "loss": 4.0894,
+      "step": 612352
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.996573881835448e-05,
+      "loss": 4.097,
+      "step": 612864
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.995735287084396e-05,
+      "loss": 4.0811,
+      "step": 613376
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.9948966923333445e-05,
+      "loss": 4.0819,
+      "step": 613888
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.9940597354626654e-05,
+      "loss": 4.0818,
+      "step": 614400
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.9932211407116134e-05,
+      "loss": 4.0877,
+      "step": 614912
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.9923825459605614e-05,
+      "loss": 4.096,
+      "step": 615424
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.991545589089882e-05,
+      "loss": 4.0822,
+      "step": 615936
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.990708632219204e-05,
+      "loss": 4.0919,
+      "step": 616448
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.989870037468151e-05,
+      "loss": 4.0874,
+      "step": 616960
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.989031442717099e-05,
+      "loss": 4.078,
+      "step": 617472
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.988192847966047e-05,
+      "loss": 4.0831,
+      "step": 617984
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.987354253214995e-05,
+      "loss": 4.0709,
+      "step": 618496
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.986515658463943e-05,
+      "loss": 4.0824,
+      "step": 619008
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.985677063712891e-05,
+      "loss": 4.075,
+      "step": 619520
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.98483846896184e-05,
+      "loss": 4.0771,
+      "step": 620032
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.983999874210788e-05,
+      "loss": 4.1035,
+      "step": 620544
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.983161279459736e-05,
+      "loss": 4.0912,
+      "step": 621056
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.982322684708684e-05,
+      "loss": 4.0823,
+      "step": 621568
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.981485727838005e-05,
+      "loss": 4.0853,
+      "step": 622080
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.980648770967326e-05,
+      "loss": 4.0923,
+      "step": 622592
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.979810176216274e-05,
+      "loss": 4.0674,
+      "step": 623104
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.9789715814652217e-05,
+      "loss": 4.0833,
+      "step": 623616
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.9781329867141697e-05,
+      "loss": 4.0841,
+      "step": 624128
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.9772943919631176e-05,
+      "loss": 4.0746,
+      "step": 624640
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.9764557972120656e-05,
+      "loss": 4.0712,
+      "step": 625152
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.9756172024610136e-05,
+      "loss": 4.0746,
+      "step": 625664
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9747786077099616e-05,
+      "loss": 4.0807,
+      "step": 626176
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.973941650839283e-05,
+      "loss": 4.089,
+      "step": 626688
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.973103056088231e-05,
+      "loss": 4.0811,
+      "step": 627200
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.972264461337179e-05,
+      "loss": 4.079,
+      "step": 627712
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9714275044665e-05,
+      "loss": 4.0836,
+      "step": 628224
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.970588909715448e-05,
+      "loss": 4.0825,
+      "step": 628736
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.969750314964396e-05,
+      "loss": 4.0818,
+      "step": 629248
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.968911720213344e-05,
+      "loss": 4.0817,
+      "step": 629760
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.968073125462292e-05,
+      "loss": 4.0669,
+      "step": 630272
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.967236168591613e-05,
+      "loss": 4.0818,
+      "step": 630784
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.966397573840561e-05,
+      "loss": 4.0684,
+      "step": 631296
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.965558979089509e-05,
+      "loss": 4.0895,
+      "step": 631808
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.964720384338457e-05,
+      "loss": 4.0826,
+      "step": 632320
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.963881789587405e-05,
+      "loss": 4.0797,
+      "step": 632832
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.963043194836353e-05,
+      "loss": 4.0859,
+      "step": 633344
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.962204600085301e-05,
+      "loss": 4.0772,
+      "step": 633856
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.961366005334249e-05,
+      "loss": 4.086,
+      "step": 634368
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.96052904846357e-05,
+      "loss": 4.0756,
+      "step": 634880
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.959690453712518e-05,
+      "loss": 4.051,
+      "step": 635392
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9588534968418395e-05,
+      "loss": 4.0927,
+      "step": 635904
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9580149020907875e-05,
+      "loss": 4.0746,
+      "step": 636416
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.957176307339735e-05,
+      "loss": 4.0913,
+      "step": 636928
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.956337712588683e-05,
+      "loss": 4.0724,
+      "step": 637440
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.955499117837631e-05,
+      "loss": 4.0698,
+      "step": 637952
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.954660523086579e-05,
+      "loss": 4.066,
+      "step": 638464
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9538235662159004e-05,
+      "loss": 4.0821,
+      "step": 638976
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9529849714648484e-05,
+      "loss": 4.0811,
+      "step": 639488
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9521463767137964e-05,
+      "loss": 4.0789,
+      "step": 640000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9513077819627444e-05,
+      "loss": 4.0911,
+      "step": 640512
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9504691872116923e-05,
+      "loss": 4.0683,
+      "step": 641024
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.949632230341013e-05,
+      "loss": 4.0657,
+      "step": 641536
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.948793635589961e-05,
+      "loss": 4.0878,
+      "step": 642048
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.947955040838909e-05,
+      "loss": 4.0624,
+      "step": 642560
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.947116446087857e-05,
+      "loss": 4.0681,
+      "step": 643072
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.946279489217178e-05,
+      "loss": 4.0717,
+      "step": 643584
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.945440894466126e-05,
+      "loss": 4.0825,
+      "step": 644096
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.944602299715074e-05,
+      "loss": 4.06,
+      "step": 644608
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.943763704964022e-05,
+      "loss": 4.0672,
+      "step": 645120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.942925110212971e-05,
+      "loss": 4.0597,
+      "step": 645632
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.942086515461919e-05,
+      "loss": 4.0729,
+      "step": 646144
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.941247920710867e-05,
+      "loss": 4.0806,
+      "step": 646656
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.940409325959815e-05,
+      "loss": 4.0727,
+      "step": 647168
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.939570731208763e-05,
+      "loss": 4.0743,
+      "step": 647680
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.938733774338084e-05,
+      "loss": 4.0706,
+      "step": 648192
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.937895179587032e-05,
+      "loss": 4.093,
+      "step": 648704
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.93705658483598e-05,
+      "loss": 4.0621,
+      "step": 649216
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.936217990084928e-05,
+      "loss": 4.0763,
+      "step": 649728
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9353810332142486e-05,
+      "loss": 4.0758,
+      "step": 650240
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9345424384631966e-05,
+      "loss": 4.0684,
+      "step": 650752
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.9337054815925175e-05,
+      "loss": 4.0774,
+      "step": 651264
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.932866886841466e-05,
+      "loss": 4.0758,
+      "step": 651776
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.932028292090414e-05,
+      "loss": 4.0725,
+      "step": 652288
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.931189697339362e-05,
+      "loss": 4.0769,
+      "step": 652800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.93035110258831e-05,
+      "loss": 4.0684,
+      "step": 653312
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.929512507837258e-05,
+      "loss": 4.0706,
+      "step": 653824
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.928675550966579e-05,
+      "loss": 4.072,
+      "step": 654336
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.927836956215527e-05,
+      "loss": 4.077,
+      "step": 654848
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.926999999344848e-05,
+      "loss": 4.0653,
+      "step": 655360
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.926161404593796e-05,
+      "loss": 4.0687,
+      "step": 655872
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.925322809842744e-05,
+      "loss": 4.0626,
+      "step": 656384
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.924484215091692e-05,
+      "loss": 4.0777,
+      "step": 656896
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.92364562034064e-05,
+      "loss": 4.0732,
+      "step": 657408
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.922807025589588e-05,
+      "loss": 4.0585,
+      "step": 657920
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.921968430838536e-05,
+      "loss": 4.0754,
+      "step": 658432
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.921129836087484e-05,
+      "loss": 4.0765,
+      "step": 658944
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.920291241336432e-05,
+      "loss": 4.0703,
+      "step": 659456
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.91945264658538e-05,
+      "loss": 4.053,
+      "step": 659968
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.918614051834328e-05,
+      "loss": 4.0651,
+      "step": 660480
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.917775457083276e-05,
+      "loss": 4.0609,
+      "step": 660992
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.916938500212597e-05,
+      "loss": 4.077,
+      "step": 661504
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.916099905461545e-05,
+      "loss": 4.0745,
+      "step": 662016
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.915261310710493e-05,
+      "loss": 4.0718,
+      "step": 662528
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.914422715959441e-05,
+      "loss": 4.065,
+      "step": 663040
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.913585759088762e-05,
+      "loss": 4.0652,
+      "step": 663552
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.91274716433771e-05,
+      "loss": 4.0598,
+      "step": 664064
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.9119085695866584e-05,
+      "loss": 4.0632,
+      "step": 664576
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.9110699748356064e-05,
+      "loss": 4.0584,
+      "step": 665088
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.910233017964927e-05,
+      "loss": 4.0757,
+      "step": 665600
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.909394423213875e-05,
+      "loss": 4.0541,
+      "step": 666112
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.908555828462823e-05,
+      "loss": 4.0562,
+      "step": 666624
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.907717233711771e-05,
+      "loss": 4.0631,
+      "step": 667136
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.906880276841092e-05,
+      "loss": 4.0617,
+      "step": 667648
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.90604168209004e-05,
+      "loss": 4.0624,
+      "step": 668160
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.905203087338988e-05,
+      "loss": 4.0577,
+      "step": 668672
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.904364492587936e-05,
+      "loss": 4.0683,
+      "step": 669184
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.903527535717257e-05,
+      "loss": 4.0733,
+      "step": 669696
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.902688940966205e-05,
+      "loss": 4.056,
+      "step": 670208
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.901851984095527e-05,
+      "loss": 4.0673,
+      "step": 670720
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.901013389344475e-05,
+      "loss": 4.0609,
+      "step": 671232
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.900174794593423e-05,
+      "loss": 4.0671,
+      "step": 671744
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.899336199842371e-05,
+      "loss": 4.0651,
+      "step": 672256
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.898497605091319e-05,
+      "loss": 4.0614,
+      "step": 672768
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8976606482206396e-05,
+      "loss": 4.0592,
+      "step": 673280
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8968220534695876e-05,
+      "loss": 4.0654,
+      "step": 673792
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8959834587185356e-05,
+      "loss": 4.0683,
+      "step": 674304
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8951448639674836e-05,
+      "loss": 4.0607,
+      "step": 674816
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8943062692164316e-05,
+      "loss": 4.0587,
+      "step": 675328
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8934693123457525e-05,
+      "loss": 4.0628,
+      "step": 675840
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8926307175947005e-05,
+      "loss": 4.0466,
+      "step": 676352
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.891792122843649e-05,
+      "loss": 4.0768,
+      "step": 676864
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.890953528092597e-05,
+      "loss": 4.0558,
+      "step": 677376
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.890114933341545e-05,
+      "loss": 4.0719,
+      "step": 677888
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.889276338590493e-05,
+      "loss": 4.0517,
+      "step": 678400
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.888437743839441e-05,
+      "loss": 4.0745,
+      "step": 678912
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.887599149088389e-05,
+      "loss": 4.068,
+      "step": 679424
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.88676219221771e-05,
+      "loss": 4.0608,
+      "step": 679936
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.885923597466658e-05,
+      "loss": 4.0606,
+      "step": 680448
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.885085002715606e-05,
+      "loss": 4.0633,
+      "step": 680960
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.884248045844927e-05,
+      "loss": 4.0659,
+      "step": 681472
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.883409451093875e-05,
+      "loss": 4.078,
+      "step": 681984
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.882570856342823e-05,
+      "loss": 4.0471,
+      "step": 682496
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.881732261591771e-05,
+      "loss": 4.0733,
+      "step": 683008
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.880893666840719e-05,
+      "loss": 4.0556,
+      "step": 683520
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.880055072089667e-05,
+      "loss": 4.0689,
+      "step": 684032
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8792181152189885e-05,
+      "loss": 4.061,
+      "step": 684544
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8783795204679365e-05,
+      "loss": 4.0564,
+      "step": 685056
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8775409257168845e-05,
+      "loss": 4.0655,
+      "step": 685568
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8767039688462054e-05,
+      "loss": 4.0658,
+      "step": 686080
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8758653740951534e-05,
+      "loss": 4.0628,
+      "step": 686592
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.136997699737549,
+      "eval_runtime": 293.1246,
+      "eval_samples_per_second": 1301.805,
+      "eval_steps_per_second": 40.682,
+      "step": 686880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8750267793441014e-05,
+      "loss": 4.056,
+      "step": 687104
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.874188184593049e-05,
+      "loss": 4.0599,
+      "step": 687616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.873349589841997e-05,
+      "loss": 4.0673,
+      "step": 688128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.872510995090945e-05,
+      "loss": 4.0622,
+      "step": 688640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.871672400339893e-05,
+      "loss": 4.0683,
+      "step": 689152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.870833805588841e-05,
+      "loss": 4.0563,
+      "step": 689664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8699952108377894e-05,
+      "loss": 4.0573,
+      "step": 690176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8691566160867374e-05,
+      "loss": 4.0569,
+      "step": 690688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8683180213356853e-05,
+      "loss": 4.0597,
+      "step": 691200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8674794265846333e-05,
+      "loss": 4.0666,
+      "step": 691712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.866640831833581e-05,
+      "loss": 4.0575,
+      "step": 692224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.865803874962902e-05,
+      "loss": 4.062,
+      "step": 692736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.86496528021185e-05,
+      "loss": 4.0641,
+      "step": 693248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.864126685460798e-05,
+      "loss": 4.0508,
+      "step": 693760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.863288090709746e-05,
+      "loss": 4.0543,
+      "step": 694272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.862449495958694e-05,
+      "loss": 4.0475,
+      "step": 694784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.861610901207642e-05,
+      "loss": 4.0525,
+      "step": 695296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.86077230645659e-05,
+      "loss": 4.0502,
+      "step": 695808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8599337117055375e-05,
+      "loss": 4.0515,
+      "step": 696320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.859095116954486e-05,
+      "loss": 4.0734,
+      "step": 696832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.858256522203434e-05,
+      "loss": 4.0663,
+      "step": 697344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.857417927452382e-05,
+      "loss": 4.0575,
+      "step": 697856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.85657933270133e-05,
+      "loss": 4.0569,
+      "step": 698368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.855740737950278e-05,
+      "loss": 4.0674,
+      "step": 698880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.854903781079599e-05,
+      "loss": 4.04,
+      "step": 699392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.854065186328547e-05,
+      "loss": 4.0553,
+      "step": 699904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.853226591577495e-05,
+      "loss": 4.0603,
+      "step": 700416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.852387996826443e-05,
+      "loss": 4.0417,
+      "step": 700928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.851549402075391e-05,
+      "loss": 4.0526,
+      "step": 701440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.850710807324339e-05,
+      "loss": 4.0486,
+      "step": 701952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.849872212573287e-05,
+      "loss": 4.0554,
+      "step": 702464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.849033617822235e-05,
+      "loss": 4.0567,
+      "step": 702976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.848196660951556e-05,
+      "loss": 4.0557,
+      "step": 703488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8473580662005047e-05,
+      "loss": 4.0611,
+      "step": 704000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8465211093298256e-05,
+      "loss": 4.0524,
+      "step": 704512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8456825145787736e-05,
+      "loss": 4.0581,
+      "step": 705024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8448439198277216e-05,
+      "loss": 4.0619,
+      "step": 705536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8440053250766696e-05,
+      "loss": 4.0514,
+      "step": 706048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8431667303256175e-05,
+      "loss": 4.0437,
+      "step": 706560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8423281355745655e-05,
+      "loss": 4.0555,
+      "step": 707072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8414895408235135e-05,
+      "loss": 4.0428,
+      "step": 707584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8406509460724615e-05,
+      "loss": 4.0643,
+      "step": 708096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8398139892017824e-05,
+      "loss": 4.059,
+      "step": 708608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8389770323311034e-05,
+      "loss": 4.0533,
+      "step": 709120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8381384375800513e-05,
+      "loss": 4.0602,
+      "step": 709632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8372998428289993e-05,
+      "loss": 4.0491,
+      "step": 710144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.836461248077948e-05,
+      "loss": 4.0606,
+      "step": 710656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.835622653326896e-05,
+      "loss": 4.0561,
+      "step": 711168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.834784058575844e-05,
+      "loss": 4.0225,
+      "step": 711680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.833945463824792e-05,
+      "loss": 4.0659,
+      "step": 712192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.833106869073739e-05,
+      "loss": 4.0511,
+      "step": 712704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.832269912203061e-05,
+      "loss": 4.0609,
+      "step": 713216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.831431317452009e-05,
+      "loss": 4.049,
+      "step": 713728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.830592722700956e-05,
+      "loss": 4.0483,
+      "step": 714240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.829754127949904e-05,
+      "loss": 4.0406,
+      "step": 714752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.828917171079226e-05,
+      "loss": 4.0552,
+      "step": 715264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.828078576328174e-05,
+      "loss": 4.0548,
+      "step": 715776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.827239981577122e-05,
+      "loss": 4.0518,
+      "step": 716288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.82640138682607e-05,
+      "loss": 4.0695,
+      "step": 716800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8255644299553914e-05,
+      "loss": 4.046,
+      "step": 717312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8247258352043394e-05,
+      "loss": 4.0352,
+      "step": 717824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.82388887833366e-05,
+      "loss": 4.0645,
+      "step": 718336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.823050283582608e-05,
+      "loss": 4.0404,
+      "step": 718848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.822211688831556e-05,
+      "loss": 4.0382,
+      "step": 719360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8213730940805036e-05,
+      "loss": 4.0556,
+      "step": 719872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8205344993294516e-05,
+      "loss": 4.055,
+      "step": 720384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8196959045783996e-05,
+      "loss": 4.04,
+      "step": 720896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8188573098273476e-05,
+      "loss": 4.0421,
+      "step": 721408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8180187150762956e-05,
+      "loss": 4.0365,
+      "step": 721920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.817181758205617e-05,
+      "loss": 4.0457,
+      "step": 722432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.816343163454565e-05,
+      "loss": 4.0575,
+      "step": 722944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.815504568703513e-05,
+      "loss": 4.0484,
+      "step": 723456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.814665973952461e-05,
+      "loss": 4.0514,
+      "step": 723968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.813829017081782e-05,
+      "loss": 4.0441,
+      "step": 724480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.81299042233073e-05,
+      "loss": 4.0663,
+      "step": 724992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.812151827579678e-05,
+      "loss": 4.0401,
+      "step": 725504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.811313232828626e-05,
+      "loss": 4.0541,
+      "step": 726016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.810476275957947e-05,
+      "loss": 4.0481,
+      "step": 726528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.809637681206895e-05,
+      "loss": 4.0447,
+      "step": 727040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.808799086455843e-05,
+      "loss": 4.0513,
+      "step": 727552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.807960491704791e-05,
+      "loss": 4.0545,
+      "step": 728064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.807121896953739e-05,
+      "loss": 4.0441,
+      "step": 728576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.806283302202687e-05,
+      "loss": 4.0557,
+      "step": 729088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8054447074516356e-05,
+      "loss": 4.0435,
+      "step": 729600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8046061127005836e-05,
+      "loss": 4.0448,
+      "step": 730112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8037691558299045e-05,
+      "loss": 4.0465,
+      "step": 730624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8029305610788525e-05,
+      "loss": 4.0504,
+      "step": 731136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8020919663278005e-05,
+      "loss": 4.0407,
+      "step": 731648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8012550094571214e-05,
+      "loss": 4.0475,
+      "step": 732160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8004164147060694e-05,
+      "loss": 4.0419,
+      "step": 732672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7995778199550174e-05,
+      "loss": 4.0484,
+      "step": 733184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7987392252039654e-05,
+      "loss": 4.0495,
+      "step": 733696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7979006304529134e-05,
+      "loss": 4.0351,
+      "step": 734208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7970620357018614e-05,
+      "loss": 4.0512,
+      "step": 734720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7962234409508094e-05,
+      "loss": 4.048,
+      "step": 735232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.795386484080131e-05,
+      "loss": 4.0513,
+      "step": 735744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.794547889329079e-05,
+      "loss": 4.0278,
+      "step": 736256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.793709294578027e-05,
+      "loss": 4.0394,
+      "step": 736768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.792870699826975e-05,
+      "loss": 4.0379,
+      "step": 737280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.792032105075922e-05,
+      "loss": 4.0544,
+      "step": 737792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.791195148205244e-05,
+      "loss": 4.0488,
+      "step": 738304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.790356553454192e-05,
+      "loss": 4.0472,
+      "step": 738816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.789519596583513e-05,
+      "loss": 4.0419,
+      "step": 739328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.788681001832461e-05,
+      "loss": 4.0386,
+      "step": 739840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.787842407081409e-05,
+      "loss": 4.0328,
+      "step": 740352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.787003812330357e-05,
+      "loss": 4.0421,
+      "step": 740864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.786165217579305e-05,
+      "loss": 4.0359,
+      "step": 741376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.785326622828253e-05,
+      "loss": 4.0475,
+      "step": 741888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.784489665957574e-05,
+      "loss": 4.0326,
+      "step": 742400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.783651071206522e-05,
+      "loss": 4.0312,
+      "step": 742912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7828124764554696e-05,
+      "loss": 4.0366,
+      "step": 743424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7819738817044176e-05,
+      "loss": 4.0403,
+      "step": 743936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7811352869533656e-05,
+      "loss": 4.0392,
+      "step": 744448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7802966922023136e-05,
+      "loss": 4.0329,
+      "step": 744960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7794580974512616e-05,
+      "loss": 4.0471,
+      "step": 745472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7786195027002096e-05,
+      "loss": 4.0486,
+      "step": 745984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7777825458295305e-05,
+      "loss": 4.032,
+      "step": 746496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.776945588958852e-05,
+      "loss": 4.0427,
+      "step": 747008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7761069942078e-05,
+      "loss": 4.0414,
+      "step": 747520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.775268399456748e-05,
+      "loss": 4.0414,
+      "step": 748032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.774429804705696e-05,
+      "loss": 4.0417,
+      "step": 748544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.773591209954644e-05,
+      "loss": 4.0357,
+      "step": 749056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.772754253083965e-05,
+      "loss": 4.0393,
+      "step": 749568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.771915658332913e-05,
+      "loss": 4.0376,
+      "step": 750080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.771077063581861e-05,
+      "loss": 4.0448,
+      "step": 750592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.770238468830809e-05,
+      "loss": 4.0385,
+      "step": 751104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.769399874079757e-05,
+      "loss": 4.0312,
+      "step": 751616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.768561279328705e-05,
+      "loss": 4.0403,
+      "step": 752128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.767722684577653e-05,
+      "loss": 4.0233,
+      "step": 752640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.766884089826601e-05,
+      "loss": 4.0528,
+      "step": 753152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.766047132955922e-05,
+      "loss": 4.0323,
+      "step": 753664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.76520853820487e-05,
+      "loss": 4.0483,
+      "step": 754176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7643715813341915e-05,
+      "loss": 4.0306,
+      "step": 754688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7635329865831395e-05,
+      "loss": 4.0481,
+      "step": 755200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7626943918320875e-05,
+      "loss": 4.0465,
+      "step": 755712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7618557970810355e-05,
+      "loss": 4.0346,
+      "step": 756224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7610188402103564e-05,
+      "loss": 4.0401,
+      "step": 756736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7601802454593044e-05,
+      "loss": 4.0364,
+      "step": 757248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7593416507082524e-05,
+      "loss": 4.0474,
+      "step": 757760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7585030559572004e-05,
+      "loss": 4.0521,
+      "step": 758272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7576644612061484e-05,
+      "loss": 4.024,
+      "step": 758784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7568258664550964e-05,
+      "loss": 4.0436,
+      "step": 759296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7559872717040443e-05,
+      "loss": 4.0375,
+      "step": 759808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.755150314833365e-05,
+      "loss": 4.0456,
+      "step": 760320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.754311720082313e-05,
+      "loss": 4.0347,
+      "step": 760832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.753473125331262e-05,
+      "loss": 4.0397,
+      "step": 761344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.75263453058021e-05,
+      "loss": 4.0404,
+      "step": 761856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.751797573709531e-05,
+      "loss": 4.0432,
+      "step": 762368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.750958978958479e-05,
+      "loss": 4.0419,
+      "step": 762880
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.12168025970459,
+      "eval_runtime": 283.8829,
+      "eval_samples_per_second": 1344.185,
+      "eval_steps_per_second": 42.007,
+      "step": 763200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.750120384207427e-05,
+      "loss": 4.0329,
+      "step": 763392
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.749281789456375e-05,
+      "loss": 4.0362,
+      "step": 763904
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.748443194705323e-05,
+      "loss": 4.043,
+      "step": 764416
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.747604599954271e-05,
+      "loss": 4.0396,
+      "step": 764928
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.746766005203218e-05,
+      "loss": 4.0441,
+      "step": 765440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.745927410452166e-05,
+      "loss": 4.0356,
+      "step": 765952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.745088815701114e-05,
+      "loss": 4.0348,
+      "step": 766464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7442534967108086e-05,
+      "loss": 4.0337,
+      "step": 766976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.743414901959757e-05,
+      "loss": 4.0356,
+      "step": 767488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.742576307208705e-05,
+      "loss": 4.0458,
+      "step": 768000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.741737712457653e-05,
+      "loss": 4.0327,
+      "step": 768512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.740900755586974e-05,
+      "loss": 4.0365,
+      "step": 769024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.740062160835922e-05,
+      "loss": 4.0433,
+      "step": 769536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.73922356608487e-05,
+      "loss": 4.0264,
+      "step": 770048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.738384971333818e-05,
+      "loss": 4.0352,
+      "step": 770560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7375463765827655e-05,
+      "loss": 4.0244,
+      "step": 771072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7367077818317135e-05,
+      "loss": 4.0295,
+      "step": 771584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7358691870806615e-05,
+      "loss": 4.0307,
+      "step": 772096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7350305923296095e-05,
+      "loss": 4.0252,
+      "step": 772608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7341919975785575e-05,
+      "loss": 4.0471,
+      "step": 773120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7333534028275055e-05,
+      "loss": 4.0472,
+      "step": 773632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.732514808076454e-05,
+      "loss": 4.0371,
+      "step": 774144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.731676213325402e-05,
+      "loss": 4.0359,
+      "step": 774656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.730840894335096e-05,
+      "loss": 4.0412,
+      "step": 775168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.730002299584044e-05,
+      "loss": 4.0241,
+      "step": 775680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.729163704832992e-05,
+      "loss": 4.0331,
+      "step": 776192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.72832511008194e-05,
+      "loss": 4.0338,
+      "step": 776704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.727486515330888e-05,
+      "loss": 4.0195,
+      "step": 777216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.726647920579836e-05,
+      "loss": 4.0273,
+      "step": 777728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.725809325828784e-05,
+      "loss": 4.0258,
+      "step": 778240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.724970731077732e-05,
+      "loss": 4.0363,
+      "step": 778752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.724133774207053e-05,
+      "loss": 4.0325,
+      "step": 779264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7232968173363744e-05,
+      "loss": 4.0353,
+      "step": 779776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7224582225853224e-05,
+      "loss": 4.0373,
+      "step": 780288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.721621265714643e-05,
+      "loss": 4.0266,
+      "step": 780800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.720782670963591e-05,
+      "loss": 4.0383,
+      "step": 781312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.719944076212539e-05,
+      "loss": 4.0419,
+      "step": 781824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.719105481461487e-05,
+      "loss": 4.0243,
+      "step": 782336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.718266886710435e-05,
+      "loss": 4.0269,
+      "step": 782848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.717428291959383e-05,
+      "loss": 4.0273,
+      "step": 783360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.716589697208331e-05,
+      "loss": 4.0248,
+      "step": 783872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.715751102457279e-05,
+      "loss": 4.0406,
+      "step": 784384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.714912507706227e-05,
+      "loss": 4.0364,
+      "step": 784896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.714073912955175e-05,
+      "loss": 4.0327,
+      "step": 785408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.713235318204123e-05,
+      "loss": 4.0357,
+      "step": 785920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.712398361333444e-05,
+      "loss": 4.03,
+      "step": 786432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.711559766582393e-05,
+      "loss": 4.0363,
+      "step": 786944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.710721171831341e-05,
+      "loss": 4.0354,
+      "step": 787456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.709882577080289e-05,
+      "loss": 4.0054,
+      "step": 787968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.709043982329237e-05,
+      "loss": 4.0422,
+      "step": 788480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.708205387578184e-05,
+      "loss": 4.0273,
+      "step": 788992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.707366792827132e-05,
+      "loss": 4.0386,
+      "step": 789504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.70652819807608e-05,
+      "loss": 4.0292,
+      "step": 790016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.705689603325028e-05,
+      "loss": 4.0254,
+      "step": 790528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.704854284334723e-05,
+      "loss": 4.0228,
+      "step": 791040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.704015689583671e-05,
+      "loss": 4.0329,
+      "step": 791552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7031770948326187e-05,
+      "loss": 4.0346,
+      "step": 792064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7023385000815667e-05,
+      "loss": 4.029,
+      "step": 792576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7014999053305147e-05,
+      "loss": 4.0472,
+      "step": 793088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7006613105794626e-05,
+      "loss": 4.0309,
+      "step": 793600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6998227158284106e-05,
+      "loss": 4.0073,
+      "step": 794112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6989841210773586e-05,
+      "loss": 4.0419,
+      "step": 794624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6981471642066795e-05,
+      "loss": 4.0198,
+      "step": 795136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6973085694556275e-05,
+      "loss": 4.0207,
+      "step": 795648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.696471612584949e-05,
+      "loss": 4.0327,
+      "step": 796160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6956330178338964e-05,
+      "loss": 4.032,
+      "step": 796672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6947944230828444e-05,
+      "loss": 4.0183,
+      "step": 797184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6939558283317924e-05,
+      "loss": 4.0213,
+      "step": 797696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6931172335807404e-05,
+      "loss": 4.0148,
+      "step": 798208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6922786388296884e-05,
+      "loss": 4.0243,
+      "step": 798720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6914400440786364e-05,
+      "loss": 4.0343,
+      "step": 799232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.690603087207958e-05,
+      "loss": 4.0285,
+      "step": 799744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.689764492456906e-05,
+      "loss": 4.0297,
+      "step": 800256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.688925897705854e-05,
+      "loss": 4.0218,
+      "step": 800768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.688087302954802e-05,
+      "loss": 4.0435,
+      "step": 801280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.687250346084123e-05,
+      "loss": 4.0237,
+      "step": 801792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.686411751333071e-05,
+      "loss": 4.0322,
+      "step": 802304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.685573156582019e-05,
+      "loss": 4.0238,
+      "step": 802816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.684734561830967e-05,
+      "loss": 4.027,
+      "step": 803328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.683895967079915e-05,
+      "loss": 4.0315,
+      "step": 803840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.683057372328863e-05,
+      "loss": 4.0334,
+      "step": 804352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.682220415458184e-05,
+      "loss": 4.0222,
+      "step": 804864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.681381820707132e-05,
+      "loss": 4.038,
+      "step": 805376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6805432259560805e-05,
+      "loss": 4.0157,
+      "step": 805888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6797046312050285e-05,
+      "loss": 4.0234,
+      "step": 806400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6788660364539765e-05,
+      "loss": 4.028,
+      "step": 806912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6780290795832974e-05,
+      "loss": 4.0333,
+      "step": 807424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6771904848322454e-05,
+      "loss": 4.0215,
+      "step": 807936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6763518900811934e-05,
+      "loss": 4.0237,
+      "step": 808448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6755132953301414e-05,
+      "loss": 4.0186,
+      "step": 808960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6746747005790893e-05,
+      "loss": 4.0245,
+      "step": 809472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.67383774370841e-05,
+      "loss": 4.0281,
+      "step": 809984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.672999148957358e-05,
+      "loss": 4.0141,
+      "step": 810496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.672160554206306e-05,
+      "loss": 4.0294,
+      "step": 811008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.671321959455254e-05,
+      "loss": 4.0304,
+      "step": 811520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.670483364704202e-05,
+      "loss": 4.0324,
+      "step": 812032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.66964476995315e-05,
+      "loss": 4.0041,
+      "step": 812544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.668806175202098e-05,
+      "loss": 4.0226,
+      "step": 813056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.667967580451046e-05,
+      "loss": 4.018,
+      "step": 813568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.667130623580368e-05,
+      "loss": 4.0302,
+      "step": 814080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.666292028829315e-05,
+      "loss": 4.0338,
+      "step": 814592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.665453434078263e-05,
+      "loss": 4.0239,
+      "step": 815104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.664614839327211e-05,
+      "loss": 4.0194,
+      "step": 815616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.663776244576159e-05,
+      "loss": 4.0225,
+      "step": 816128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.66293928770548e-05,
+      "loss": 4.0161,
+      "step": 816640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6621023308348016e-05,
+      "loss": 4.0209,
+      "step": 817152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6612637360837496e-05,
+      "loss": 4.0174,
+      "step": 817664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6604251413326976e-05,
+      "loss": 4.0262,
+      "step": 818176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6595865465816456e-05,
+      "loss": 4.0134,
+      "step": 818688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6587479518305936e-05,
+      "loss": 4.0127,
+      "step": 819200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6579093570795416e-05,
+      "loss": 4.0143,
+      "step": 819712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6570707623284896e-05,
+      "loss": 4.0174,
+      "step": 820224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6562338054578105e-05,
+      "loss": 4.0221,
+      "step": 820736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6553952107067585e-05,
+      "loss": 4.0129,
+      "step": 821248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6545566159557065e-05,
+      "loss": 4.0244,
+      "step": 821760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6537180212046545e-05,
+      "loss": 4.0257,
+      "step": 822272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6528794264536025e-05,
+      "loss": 4.0171,
+      "step": 822784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6520424695829234e-05,
+      "loss": 4.0249,
+      "step": 823296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6512038748318714e-05,
+      "loss": 4.02,
+      "step": 823808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.650366917961193e-05,
+      "loss": 4.0216,
+      "step": 824320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.649528323210141e-05,
+      "loss": 4.0185,
+      "step": 824832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.648689728459089e-05,
+      "loss": 4.016,
+      "step": 825344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.647851133708037e-05,
+      "loss": 4.0221,
+      "step": 825856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.647012538956985e-05,
+      "loss": 4.0153,
+      "step": 826368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.646173944205933e-05,
+      "loss": 4.0267,
+      "step": 826880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.645335349454881e-05,
+      "loss": 4.0226,
+      "step": 827392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.644496754703829e-05,
+      "loss": 4.0122,
+      "step": 827904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.64365979783315e-05,
+      "loss": 4.016,
+      "step": 828416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.642821203082098e-05,
+      "loss": 4.0062,
+      "step": 828928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.641982608331046e-05,
+      "loss": 4.031,
+      "step": 829440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.641144013579994e-05,
+      "loss": 4.0122,
+      "step": 829952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.640305418828942e-05,
+      "loss": 4.0281,
+      "step": 830464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.63946682407789e-05,
+      "loss": 4.0148,
+      "step": 830976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6386298672072114e-05,
+      "loss": 4.0272,
+      "step": 831488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.637792910336532e-05,
+      "loss": 4.026,
+      "step": 832000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.63695431558548e-05,
+      "loss": 4.0133,
+      "step": 832512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.636115720834428e-05,
+      "loss": 4.0231,
+      "step": 833024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.635277126083376e-05,
+      "loss": 4.0153,
+      "step": 833536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.634438531332324e-05,
+      "loss": 4.0293,
+      "step": 834048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.633599936581272e-05,
+      "loss": 4.0254,
+      "step": 834560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.63276134183022e-05,
+      "loss": 4.0127,
+      "step": 835072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.631922747079168e-05,
+      "loss": 4.0203,
+      "step": 835584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6310841523281156e-05,
+      "loss": 4.0199,
+      "step": 836096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6302455575770636e-05,
+      "loss": 4.0266,
+      "step": 836608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.629408600706385e-05,
+      "loss": 4.0127,
+      "step": 837120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.628570005955333e-05,
+      "loss": 4.0231,
+      "step": 837632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.627733049084655e-05,
+      "loss": 4.0145,
+      "step": 838144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.626894454333603e-05,
+      "loss": 4.0265,
+      "step": 838656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.626055859582551e-05,
+      "loss": 4.0191,
+      "step": 839168
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.109185695648193,
+      "eval_runtime": 304.8366,
+      "eval_samples_per_second": 1251.789,
+      "eval_steps_per_second": 39.119,
+      "step": 839520
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.625217264831499e-05,
+      "loss": 4.0148,
+      "step": 839680
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.624378670080446e-05,
+      "loss": 4.0148,
+      "step": 840192
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.623540075329394e-05,
+      "loss": 4.022,
+      "step": 840704
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.622701480578342e-05,
+      "loss": 4.0244,
+      "step": 841216
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.62186288582729e-05,
+      "loss": 4.0262,
+      "step": 841728
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.621024291076238e-05,
+      "loss": 4.0151,
+      "step": 842240
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.620185696325186e-05,
+      "loss": 4.0144,
+      "step": 842752
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.619347101574134e-05,
+      "loss": 4.0157,
+      "step": 843264
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.618508506823082e-05,
+      "loss": 4.0187,
+      "step": 843776
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6176715499524036e-05,
+      "loss": 4.0214,
+      "step": 844288
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6168329552013516e-05,
+      "loss": 4.0131,
+      "step": 844800
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6159959983306725e-05,
+      "loss": 4.0179,
+      "step": 845312
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6151574035796205e-05,
+      "loss": 4.0282,
+      "step": 845824
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6143188088285685e-05,
+      "loss": 4.0078,
+      "step": 846336
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6134802140775165e-05,
+      "loss": 4.0107,
+      "step": 846848
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6126416193264645e-05,
+      "loss": 4.0072,
+      "step": 847360
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6118030245754125e-05,
+      "loss": 4.0068,
+      "step": 847872
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6109644298243605e-05,
+      "loss": 4.0134,
+      "step": 848384
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6101258350733085e-05,
+      "loss": 4.0113,
+      "step": 848896
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6092888782026294e-05,
+      "loss": 4.0241,
+      "step": 849408
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.60845192133195e-05,
+      "loss": 4.0274,
+      "step": 849920
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.607613326580899e-05,
+      "loss": 4.0163,
+      "step": 850432
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.606774731829847e-05,
+      "loss": 4.018,
+      "step": 850944
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.605936137078795e-05,
+      "loss": 4.0221,
+      "step": 851456
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.605099180208116e-05,
+      "loss": 4.0029,
+      "step": 851968
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.604260585457064e-05,
+      "loss": 4.0185,
+      "step": 852480
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.603421990706012e-05,
+      "loss": 4.0134,
+      "step": 852992
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.60258339595496e-05,
+      "loss": 4.0014,
+      "step": 853504
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.601744801203908e-05,
+      "loss": 4.0094,
+      "step": 854016
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.600906206452856e-05,
+      "loss": 4.0097,
+      "step": 854528
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.600069249582177e-05,
+      "loss": 4.0115,
+      "step": 855040
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.599230654831125e-05,
+      "loss": 4.0135,
+      "step": 855552
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.598392060080073e-05,
+      "loss": 4.0183,
+      "step": 856064
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.597553465329021e-05,
+      "loss": 4.0139,
+      "step": 856576
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.596714870577969e-05,
+      "loss": 4.0051,
+      "step": 857088
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5958779137072904e-05,
+      "loss": 4.0206,
+      "step": 857600
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5950393189562384e-05,
+      "loss": 4.0221,
+      "step": 858112
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5942007242051864e-05,
+      "loss": 4.007,
+      "step": 858624
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5933621294541344e-05,
+      "loss": 4.0091,
+      "step": 859136
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.592525172583455e-05,
+      "loss": 4.0077,
+      "step": 859648
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.591686577832403e-05,
+      "loss": 4.0026,
+      "step": 860160
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.590847983081351e-05,
+      "loss": 4.0199,
+      "step": 860672
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.590009388330299e-05,
+      "loss": 4.0172,
+      "step": 861184
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.58917243145962e-05,
+      "loss": 4.0201,
+      "step": 861696
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.588333836708568e-05,
+      "loss": 4.0124,
+      "step": 862208
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.587495241957516e-05,
+      "loss": 4.0141,
+      "step": 862720
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.586656647206464e-05,
+      "loss": 4.0161,
+      "step": 863232
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.585818052455412e-05,
+      "loss": 4.0153,
+      "step": 863744
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.58497945770436e-05,
+      "loss": 3.9882,
+      "step": 864256
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.584142500833682e-05,
+      "loss": 4.0223,
+      "step": 864768
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.583303906082629e-05,
+      "loss": 4.0076,
+      "step": 865280
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.582465311331577e-05,
+      "loss": 4.0231,
+      "step": 865792
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.581626716580525e-05,
+      "loss": 4.0104,
+      "step": 866304
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.580788121829473e-05,
+      "loss": 4.0083,
+      "step": 866816
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.579949527078421e-05,
+      "loss": 4.0021,
+      "step": 867328
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.579110932327369e-05,
+      "loss": 4.0113,
+      "step": 867840
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.578272337576317e-05,
+      "loss": 4.0212,
+      "step": 868352
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.577435380705638e-05,
+      "loss": 4.0073,
+      "step": 868864
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.576596785954586e-05,
+      "loss": 4.0255,
+      "step": 869376
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5757581912035346e-05,
+      "loss": 4.0115,
+      "step": 869888
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5749195964524826e-05,
+      "loss": 3.9904,
+      "step": 870400
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5740810017014306e-05,
+      "loss": 4.0231,
+      "step": 870912
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5732440448307515e-05,
+      "loss": 4.0028,
+      "step": 871424
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5724070879600724e-05,
+      "loss": 4.0045,
+      "step": 871936
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5715684932090204e-05,
+      "loss": 4.0103,
+      "step": 872448
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5707298984579684e-05,
+      "loss": 4.0154,
+      "step": 872960
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5698913037069164e-05,
+      "loss": 3.9977,
+      "step": 873472
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5690527089558644e-05,
+      "loss": 4.003,
+      "step": 873984
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5682141142048124e-05,
+      "loss": 3.9947,
+      "step": 874496
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5673755194537604e-05,
+      "loss": 4.0107,
+      "step": 875008
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5665369247027084e-05,
+      "loss": 4.0123,
+      "step": 875520
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.56569996783203e-05,
+      "loss": 4.0082,
+      "step": 876032
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.564861373080978e-05,
+      "loss": 4.0133,
+      "step": 876544
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.564022778329926e-05,
+      "loss": 4.0019,
+      "step": 877056
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.563184183578874e-05,
+      "loss": 4.0221,
+      "step": 877568
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.562347226708195e-05,
+      "loss": 4.0094,
+      "step": 878080
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.561508631957143e-05,
+      "loss": 4.0154,
+      "step": 878592
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.560670037206091e-05,
+      "loss": 4.0055,
+      "step": 879104
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.559831442455039e-05,
+      "loss": 4.009,
+      "step": 879616
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.55899448558436e-05,
+      "loss": 4.0107,
+      "step": 880128
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.558155890833308e-05,
+      "loss": 4.0135,
+      "step": 880640
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.557317296082256e-05,
+      "loss": 4.0077,
+      "step": 881152
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.556478701331204e-05,
+      "loss": 4.0168,
+      "step": 881664
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.555640106580152e-05,
+      "loss": 3.9972,
+      "step": 882176
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5548015118291e-05,
+      "loss": 4.0051,
+      "step": 882688
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.553962917078048e-05,
+      "loss": 4.0066,
+      "step": 883200
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.553124322326996e-05,
+      "loss": 4.0165,
+      "step": 883712
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.552285727575944e-05,
+      "loss": 4.0005,
+      "step": 884224
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.551448770705265e-05,
+      "loss": 4.0046,
+      "step": 884736
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.550611813834586e-05,
+      "loss": 4.0046,
+      "step": 885248
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.549773219083534e-05,
+      "loss": 4.0061,
+      "step": 885760
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.548934624332482e-05,
+      "loss": 4.0093,
+      "step": 886272
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.54809602958143e-05,
+      "loss": 3.997,
+      "step": 886784
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5472574348303775e-05,
+      "loss": 4.0096,
+      "step": 887296
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5464188400793255e-05,
+      "loss": 4.0132,
+      "step": 887808
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.545581883208647e-05,
+      "loss": 4.0146,
+      "step": 888320
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.544744926337969e-05,
+      "loss": 3.9877,
+      "step": 888832
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.543906331586917e-05,
+      "loss": 4.0022,
+      "step": 889344
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.543067736835865e-05,
+      "loss": 3.9978,
+      "step": 889856
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.542229142084813e-05,
+      "loss": 4.0114,
+      "step": 890368
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.54139054733376e-05,
+      "loss": 4.0153,
+      "step": 890880
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.540551952582708e-05,
+      "loss": 4.005,
+      "step": 891392
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.539713357831656e-05,
+      "loss": 4.0027,
+      "step": 891904
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.538874763080604e-05,
+      "loss": 4.0039,
+      "step": 892416
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.538036168329552e-05,
+      "loss": 3.9963,
+      "step": 892928
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.537199211458873e-05,
+      "loss": 4.0051,
+      "step": 893440
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.536360616707821e-05,
+      "loss": 3.9998,
+      "step": 893952
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.535522021956769e-05,
+      "loss": 4.0068,
+      "step": 894464
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5346834272057175e-05,
+      "loss": 3.9988,
+      "step": 894976
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5338448324546655e-05,
+      "loss": 3.9951,
+      "step": 895488
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5330062377036135e-05,
+      "loss": 3.9963,
+      "step": 896000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5321676429525615e-05,
+      "loss": 3.9966,
+      "step": 896512
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5313290482015095e-05,
+      "loss": 4.0051,
+      "step": 897024
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5304920913308304e-05,
+      "loss": 3.9962,
+      "step": 897536
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5296534965797784e-05,
+      "loss": 4.0074,
+      "step": 898048
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5288149018287264e-05,
+      "loss": 4.0038,
+      "step": 898560
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5279763070776744e-05,
+      "loss": 4.0011,
+      "step": 899072
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.527139350206995e-05,
+      "loss": 4.006,
+      "step": 899584
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.526302393336316e-05,
+      "loss": 3.9988,
+      "step": 900096
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.525465436465638e-05,
+      "loss": 4.0063,
+      "step": 900608
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.524626841714586e-05,
+      "loss": 3.9988,
+      "step": 901120
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.523788246963534e-05,
+      "loss": 3.9975,
+      "step": 901632
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.522949652212482e-05,
+      "loss": 4.0058,
+      "step": 902144
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.52211105746143e-05,
+      "loss": 4.0004,
+      "step": 902656
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.521274100590751e-05,
+      "loss": 4.0065,
+      "step": 903168
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.520435505839699e-05,
+      "loss": 3.9994,
+      "step": 903680
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.519596911088647e-05,
+      "loss": 3.9949,
+      "step": 904192
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.518758316337595e-05,
+      "loss": 3.9975,
+      "step": 904704
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.517919721586543e-05,
+      "loss": 3.99,
+      "step": 905216
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.517081126835491e-05,
+      "loss": 4.0103,
+      "step": 905728
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.516242532084439e-05,
+      "loss": 3.9986,
+      "step": 906240
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.515403937333387e-05,
+      "loss": 4.0101,
+      "step": 906752
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.514565342582335e-05,
+      "loss": 3.9951,
+      "step": 907264
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.513726747831283e-05,
+      "loss": 4.0069,
+      "step": 907776
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.512888153080231e-05,
+      "loss": 4.0086,
+      "step": 908288
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.512051196209552e-05,
+      "loss": 3.9981,
+      "step": 908800
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5112126014585e-05,
+      "loss": 4.0004,
+      "step": 909312
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.510374006707448e-05,
+      "loss": 3.9982,
+      "step": 909824
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.509535411956396e-05,
+      "loss": 4.0102,
+      "step": 910336
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5086968172053436e-05,
+      "loss": 4.0102,
+      "step": 910848
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5078582224542916e-05,
+      "loss": 3.9979,
+      "step": 911360
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5070196277032396e-05,
+      "loss": 3.9961,
+      "step": 911872
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5061810329521876e-05,
+      "loss": 4.006,
+      "step": 912384
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5053440760815085e-05,
+      "loss": 4.0088,
+      "step": 912896
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.50450711921083e-05,
+      "loss": 3.9928,
+      "step": 913408
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.503668524459778e-05,
+      "loss": 4.006,
+      "step": 913920
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.502829929708726e-05,
+      "loss": 3.9972,
+      "step": 914432
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.501991334957674e-05,
+      "loss": 4.0093,
+      "step": 914944
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.501152740206622e-05,
+      "loss": 4.0021,
+      "step": 915456
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.097940444946289,
+      "eval_runtime": 306.8756,
+      "eval_samples_per_second": 1243.471,
+      "eval_steps_per_second": 38.859,
+      "step": 915840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.50031414545557e-05,
+      "loss": 4.0043,
+      "step": 915968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.499475550704518e-05,
+      "loss": 3.9936,
+      "step": 916480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.498636955953466e-05,
+      "loss": 4.0008,
+      "step": 916992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.497798361202414e-05,
+      "loss": 4.0057,
+      "step": 917504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.496959766451362e-05,
+      "loss": 4.0077,
+      "step": 918016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.49612117170031e-05,
+      "loss": 4.0003,
+      "step": 918528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.495282576949258e-05,
+      "loss": 3.9956,
+      "step": 919040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.494443982198206e-05,
+      "loss": 3.9994,
+      "step": 919552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.493605387447154e-05,
+      "loss": 3.9984,
+      "step": 920064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.492766792696102e-05,
+      "loss": 4.0016,
+      "step": 920576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.49192819794505e-05,
+      "loss": 3.9989,
+      "step": 921088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.491089603193998e-05,
+      "loss": 4.0007,
+      "step": 921600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4902526463233196e-05,
+      "loss": 4.0114,
+      "step": 922112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4894140515722676e-05,
+      "loss": 3.9888,
+      "step": 922624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.488575456821215e-05,
+      "loss": 3.9928,
+      "step": 923136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.487736862070163e-05,
+      "loss": 3.9903,
+      "step": 923648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.486898267319111e-05,
+      "loss": 3.9886,
+      "step": 924160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.486059672568059e-05,
+      "loss": 3.9966,
+      "step": 924672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.485221077817007e-05,
+      "loss": 3.9919,
+      "step": 925184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.484384120946328e-05,
+      "loss": 4.0052,
+      "step": 925696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.483545526195276e-05,
+      "loss": 4.0119,
+      "step": 926208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.482706931444224e-05,
+      "loss": 3.9987,
+      "step": 926720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.481868336693172e-05,
+      "loss": 4.0003,
+      "step": 927232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.48102974194212e-05,
+      "loss": 4.0056,
+      "step": 927744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.480191147191068e-05,
+      "loss": 3.9825,
+      "step": 928256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4793525524400164e-05,
+      "loss": 4.0038,
+      "step": 928768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4785139576889644e-05,
+      "loss": 3.9933,
+      "step": 929280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4776753629379124e-05,
+      "loss": 3.9872,
+      "step": 929792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4768367681868604e-05,
+      "loss": 3.993,
+      "step": 930304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4759981734358084e-05,
+      "loss": 3.9906,
+      "step": 930816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4751595786847564e-05,
+      "loss": 3.994,
+      "step": 931328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.47432425969445e-05,
+      "loss": 3.9943,
+      "step": 931840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.473485664943398e-05,
+      "loss": 4.0051,
+      "step": 932352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.472647070192346e-05,
+      "loss": 3.9955,
+      "step": 932864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.471808475441294e-05,
+      "loss": 3.9911,
+      "step": 933376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.470971518570615e-05,
+      "loss": 4.0002,
+      "step": 933888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.470132923819563e-05,
+      "loss": 4.0052,
+      "step": 934400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.469294329068512e-05,
+      "loss": 3.9914,
+      "step": 934912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.46845573431746e-05,
+      "loss": 3.989,
+      "step": 935424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.467617139566408e-05,
+      "loss": 3.9923,
+      "step": 935936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.466778544815356e-05,
+      "loss": 3.9882,
+      "step": 936448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.465939950064304e-05,
+      "loss": 3.9986,
+      "step": 936960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.465101355313251e-05,
+      "loss": 3.9981,
+      "step": 937472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.464262760562199e-05,
+      "loss": 4.0048,
+      "step": 937984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.463424165811147e-05,
+      "loss": 3.9968,
+      "step": 938496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.462585571060095e-05,
+      "loss": 3.9954,
+      "step": 939008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.461746976309043e-05,
+      "loss": 3.9998,
+      "step": 939520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.460908381557991e-05,
+      "loss": 3.9973,
+      "step": 940032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.460071424687312e-05,
+      "loss": 3.9715,
+      "step": 940544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.45923282993626e-05,
+      "loss": 4.0016,
+      "step": 941056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4583942351852087e-05,
+      "loss": 3.9929,
+      "step": 941568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4575572783145296e-05,
+      "loss": 4.0061,
+      "step": 942080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4567186835634776e-05,
+      "loss": 3.99,
+      "step": 942592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4558800888124256e-05,
+      "loss": 3.9945,
+      "step": 943104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4550414940613735e-05,
+      "loss": 3.9818,
+      "step": 943616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4542028993103215e-05,
+      "loss": 3.9899,
+      "step": 944128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4533643045592695e-05,
+      "loss": 4.009,
+      "step": 944640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4525257098082175e-05,
+      "loss": 3.9904,
+      "step": 945152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4516871150571655e-05,
+      "loss": 4.0071,
+      "step": 945664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4508485203061135e-05,
+      "loss": 3.9962,
+      "step": 946176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4500115634354344e-05,
+      "loss": 3.9722,
+      "step": 946688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4491729686843824e-05,
+      "loss": 4.0108,
+      "step": 947200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4483343739333304e-05,
+      "loss": 3.9827,
+      "step": 947712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4474957791822784e-05,
+      "loss": 3.9851,
+      "step": 948224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4466588223116e-05,
+      "loss": 3.9957,
+      "step": 948736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.445820227560548e-05,
+      "loss": 3.996,
+      "step": 949248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.444981632809496e-05,
+      "loss": 3.9842,
+      "step": 949760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.444143038058444e-05,
+      "loss": 3.9849,
+      "step": 950272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.443306081187765e-05,
+      "loss": 3.9803,
+      "step": 950784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.442467486436713e-05,
+      "loss": 3.9853,
+      "step": 951296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.441628891685661e-05,
+      "loss": 4.0008,
+      "step": 951808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.440790296934609e-05,
+      "loss": 3.9948,
+      "step": 952320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.43995334006393e-05,
+      "loss": 3.994,
+      "step": 952832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.439114745312878e-05,
+      "loss": 3.9853,
+      "step": 953344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.438276150561826e-05,
+      "loss": 4.0059,
+      "step": 953856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.437437555810774e-05,
+      "loss": 3.993,
+      "step": 954368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4366005989400954e-05,
+      "loss": 3.9973,
+      "step": 954880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4357620041890434e-05,
+      "loss": 3.9885,
+      "step": 955392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4349234094379914e-05,
+      "loss": 3.9924,
+      "step": 955904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.434086452567312e-05,
+      "loss": 3.9973,
+      "step": 956416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.43324785781626e-05,
+      "loss": 3.9944,
+      "step": 956928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.432409263065208e-05,
+      "loss": 3.9932,
+      "step": 957440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.431570668314156e-05,
+      "loss": 4.001,
+      "step": 957952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.430732073563104e-05,
+      "loss": 3.9831,
+      "step": 958464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4298934788120516e-05,
+      "loss": 3.9862,
+      "step": 958976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.429056521941373e-05,
+      "loss": 3.992,
+      "step": 959488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.428217927190321e-05,
+      "loss": 3.9979,
+      "step": 960000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.427379332439269e-05,
+      "loss": 3.9849,
+      "step": 960512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.426540737688217e-05,
+      "loss": 3.9915,
+      "step": 961024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.425702142937165e-05,
+      "loss": 3.9866,
+      "step": 961536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.424863548186113e-05,
+      "loss": 3.9893,
+      "step": 962048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.424026591315435e-05,
+      "loss": 3.9936,
+      "step": 962560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.423187996564382e-05,
+      "loss": 3.9825,
+      "step": 963072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.42234940181333e-05,
+      "loss": 3.9924,
+      "step": 963584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.421510807062278e-05,
+      "loss": 3.9963,
+      "step": 964096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.420672212311226e-05,
+      "loss": 4.0015,
+      "step": 964608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.419835255440547e-05,
+      "loss": 3.9675,
+      "step": 965120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.418996660689495e-05,
+      "loss": 3.987,
+      "step": 965632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.418158065938443e-05,
+      "loss": 3.9833,
+      "step": 966144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.417319471187391e-05,
+      "loss": 3.9968,
+      "step": 966656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4164808764363396e-05,
+      "loss": 3.9987,
+      "step": 967168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4156422816852876e-05,
+      "loss": 3.9889,
+      "step": 967680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4148036869342356e-05,
+      "loss": 3.9865,
+      "step": 968192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4139650921831836e-05,
+      "loss": 3.9893,
+      "step": 968704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4131264974321316e-05,
+      "loss": 3.983,
+      "step": 969216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4122895405614525e-05,
+      "loss": 3.9858,
+      "step": 969728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4114525836907734e-05,
+      "loss": 3.9819,
+      "step": 970240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4106139889397214e-05,
+      "loss": 3.9912,
+      "step": 970752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4097753941886694e-05,
+      "loss": 3.9809,
+      "step": 971264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4089367994376174e-05,
+      "loss": 3.9784,
+      "step": 971776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4080982046865654e-05,
+      "loss": 3.9813,
+      "step": 972288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4072596099355134e-05,
+      "loss": 3.9781,
+      "step": 972800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4064210151844614e-05,
+      "loss": 3.9914,
+      "step": 973312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4055824204334094e-05,
+      "loss": 3.9761,
+      "step": 973824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.404743825682358e-05,
+      "loss": 3.9931,
+      "step": 974336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.403906868811679e-05,
+      "loss": 3.99,
+      "step": 974848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.403068274060627e-05,
+      "loss": 3.9869,
+      "step": 975360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.402231317189948e-05,
+      "loss": 3.9874,
+      "step": 975872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.401392722438896e-05,
+      "loss": 3.9852,
+      "step": 976384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.400554127687844e-05,
+      "loss": 3.9921,
+      "step": 976896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.399715532936792e-05,
+      "loss": 3.9835,
+      "step": 977408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.398878576066113e-05,
+      "loss": 3.9806,
+      "step": 977920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.398039981315061e-05,
+      "loss": 3.9883,
+      "step": 978432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.397203024444382e-05,
+      "loss": 3.9843,
+      "step": 978944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.39636442969333e-05,
+      "loss": 3.989,
+      "step": 979456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.395525834942278e-05,
+      "loss": 3.9829,
+      "step": 979968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.394687240191226e-05,
+      "loss": 3.9804,
+      "step": 980480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.393848645440174e-05,
+      "loss": 3.9821,
+      "step": 980992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.393010050689122e-05,
+      "loss": 3.9783,
+      "step": 981504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.39217145593807e-05,
+      "loss": 3.9882,
+      "step": 982016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.391332861187018e-05,
+      "loss": 3.9864,
+      "step": 982528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3904942664359656e-05,
+      "loss": 3.9972,
+      "step": 983040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3896556716849136e-05,
+      "loss": 3.98,
+      "step": 983552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3888170769338616e-05,
+      "loss": 3.9873,
+      "step": 984064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3879784821828096e-05,
+      "loss": 3.9946,
+      "step": 984576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3871415253121305e-05,
+      "loss": 3.9855,
+      "step": 985088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3863029305610785e-05,
+      "loss": 3.982,
+      "step": 985600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.385464335810027e-05,
+      "loss": 3.9832,
+      "step": 986112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.384625741058975e-05,
+      "loss": 3.9956,
+      "step": 986624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.383788784188296e-05,
+      "loss": 3.9935,
+      "step": 987136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.382950189437244e-05,
+      "loss": 3.9842,
+      "step": 987648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.382111594686192e-05,
+      "loss": 3.9793,
+      "step": 988160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.38127299993514e-05,
+      "loss": 3.9906,
+      "step": 988672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.380436043064461e-05,
+      "loss": 3.9934,
+      "step": 989184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3795990861937826e-05,
+      "loss": 3.9758,
+      "step": 989696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.37876049144273e-05,
+      "loss": 3.9996,
+      "step": 990208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.377921896691678e-05,
+      "loss": 3.9775,
+      "step": 990720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.377083301940626e-05,
+      "loss": 3.9933,
+      "step": 991232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.376244707189574e-05,
+      "loss": 3.9884,
+      "step": 991744
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.0884881019592285,
+      "eval_runtime": 289.5522,
+      "eval_samples_per_second": 1317.866,
+      "eval_steps_per_second": 41.184,
+      "step": 992160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3754061124385226e-05,
+      "loss": 3.9878,
+      "step": 992256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3745691555678435e-05,
+      "loss": 3.9795,
+      "step": 992768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3737305608167915e-05,
+      "loss": 3.9852,
+      "step": 993280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3728919660657395e-05,
+      "loss": 3.9944,
+      "step": 993792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3720533713146875e-05,
+      "loss": 3.9871,
+      "step": 994304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3712147765636355e-05,
+      "loss": 3.9883,
+      "step": 994816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3703778196929564e-05,
+      "loss": 3.9807,
+      "step": 995328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3695392249419044e-05,
+      "loss": 3.9812,
+      "step": 995840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3687006301908524e-05,
+      "loss": 3.9846,
+      "step": 996352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3678620354398003e-05,
+      "loss": 3.9908,
+      "step": 996864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3670234406887483e-05,
+      "loss": 3.9817,
+      "step": 997376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3661848459376963e-05,
+      "loss": 3.992,
+      "step": 997888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.365349526947391e-05,
+      "loss": 3.9939,
+      "step": 998400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.364510932196339e-05,
+      "loss": 3.9718,
+      "step": 998912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.363672337445287e-05,
+      "loss": 3.9823,
+      "step": 999424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.362833742694235e-05,
+      "loss": 3.9693,
+      "step": 999936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.361995147943183e-05,
+      "loss": 3.9717,
+      "step": 1000448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.361156553192131e-05,
+      "loss": 3.9853,
+      "step": 1000960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.360317958441079e-05,
+      "loss": 3.9745,
+      "step": 1001472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.359479363690027e-05,
+      "loss": 3.9892,
+      "step": 1001984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.358640768938975e-05,
+      "loss": 3.9961,
+      "step": 1002496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.357802174187923e-05,
+      "loss": 3.9825,
+      "step": 1003008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.356963579436871e-05,
+      "loss": 3.9875,
+      "step": 1003520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.356124984685819e-05,
+      "loss": 3.9905,
+      "step": 1004032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.35528802781514e-05,
+      "loss": 3.9692,
+      "step": 1004544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.354449433064088e-05,
+      "loss": 3.9848,
+      "step": 1005056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.353610838313036e-05,
+      "loss": 3.9782,
+      "step": 1005568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3527722435619844e-05,
+      "loss": 3.9733,
+      "step": 1006080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.351933648810932e-05,
+      "loss": 3.9801,
+      "step": 1006592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.351098329820626e-05,
+      "loss": 3.9721,
+      "step": 1007104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.350259735069574e-05,
+      "loss": 3.9786,
+      "step": 1007616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.349421140318522e-05,
+      "loss": 3.9817,
+      "step": 1008128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.34858254556747e-05,
+      "loss": 3.9889,
+      "step": 1008640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.347743950816418e-05,
+      "loss": 3.9803,
+      "step": 1009152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.346905356065366e-05,
+      "loss": 3.9771,
+      "step": 1009664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.346068399194687e-05,
+      "loss": 3.9801,
+      "step": 1010176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.345229804443635e-05,
+      "loss": 3.9915,
+      "step": 1010688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.344391209692583e-05,
+      "loss": 3.978,
+      "step": 1011200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.343552614941531e-05,
+      "loss": 3.9736,
+      "step": 1011712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3427156580708526e-05,
+      "loss": 3.9774,
+      "step": 1012224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3418770633198006e-05,
+      "loss": 3.978,
+      "step": 1012736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3410384685687486e-05,
+      "loss": 3.9783,
+      "step": 1013248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.340199873817696e-05,
+      "loss": 3.9867,
+      "step": 1013760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.339361279066644e-05,
+      "loss": 3.9873,
+      "step": 1014272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.338522684315592e-05,
+      "loss": 3.9834,
+      "step": 1014784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3376873653252864e-05,
+      "loss": 3.9794,
+      "step": 1015296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3368487705742344e-05,
+      "loss": 3.9819,
+      "step": 1015808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3360101758231824e-05,
+      "loss": 3.9817,
+      "step": 1016320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3351715810721304e-05,
+      "loss": 3.9637,
+      "step": 1016832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3343329863210784e-05,
+      "loss": 3.9864,
+      "step": 1017344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3334943915700264e-05,
+      "loss": 3.9806,
+      "step": 1017856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3326557968189744e-05,
+      "loss": 3.987,
+      "step": 1018368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3318172020679224e-05,
+      "loss": 3.9759,
+      "step": 1018880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3309786073168704e-05,
+      "loss": 3.9772,
+      "step": 1019392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3301400125658184e-05,
+      "loss": 3.972,
+      "step": 1019904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3293014178147664e-05,
+      "loss": 3.9703,
+      "step": 1020416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.328464460944087e-05,
+      "loss": 3.9932,
+      "step": 1020928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.327625866193035e-05,
+      "loss": 3.9747,
+      "step": 1021440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.326787271441983e-05,
+      "loss": 3.9981,
+      "step": 1021952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.325948676690931e-05,
+      "loss": 3.9759,
+      "step": 1022464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.325110081939879e-05,
+      "loss": 3.962,
+      "step": 1022976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.324271487188827e-05,
+      "loss": 3.995,
+      "step": 1023488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.323434530318149e-05,
+      "loss": 3.9718,
+      "step": 1024000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.322595935567097e-05,
+      "loss": 3.9682,
+      "step": 1024512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.321757340816045e-05,
+      "loss": 3.9765,
+      "step": 1025024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.320918746064993e-05,
+      "loss": 3.9862,
+      "step": 1025536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.320080151313941e-05,
+      "loss": 3.9654,
+      "step": 1026048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.319241556562889e-05,
+      "loss": 3.9726,
+      "step": 1026560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.318402961811837e-05,
+      "loss": 3.9666,
+      "step": 1027072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.317564367060785e-05,
+      "loss": 3.9666,
+      "step": 1027584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.316727410190106e-05,
+      "loss": 3.9869,
+      "step": 1028096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.315888815439054e-05,
+      "loss": 3.979,
+      "step": 1028608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.315050220688002e-05,
+      "loss": 3.9787,
+      "step": 1029120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.31421162593695e-05,
+      "loss": 3.9702,
+      "step": 1029632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.313373031185897e-05,
+      "loss": 3.9894,
+      "step": 1030144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3125360743152187e-05,
+      "loss": 3.98,
+      "step": 1030656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3116974795641666e-05,
+      "loss": 3.9851,
+      "step": 1031168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3108588848131146e-05,
+      "loss": 3.9731,
+      "step": 1031680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3100202900620626e-05,
+      "loss": 3.9761,
+      "step": 1032192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.309183333191384e-05,
+      "loss": 3.9811,
+      "step": 1032704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.308344738440332e-05,
+      "loss": 3.9775,
+      "step": 1033216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3075061436892795e-05,
+      "loss": 3.9778,
+      "step": 1033728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.306669186818601e-05,
+      "loss": 3.9842,
+      "step": 1034240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.305830592067549e-05,
+      "loss": 3.975,
+      "step": 1034752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.304991997316497e-05,
+      "loss": 3.9676,
+      "step": 1035264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3041534025654444e-05,
+      "loss": 3.9738,
+      "step": 1035776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3033148078143924e-05,
+      "loss": 3.9813,
+      "step": 1036288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.302477850943714e-05,
+      "loss": 3.9734,
+      "step": 1036800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3016408940730356e-05,
+      "loss": 3.9738,
+      "step": 1037312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3008022993219836e-05,
+      "loss": 3.9718,
+      "step": 1037824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2999637045709316e-05,
+      "loss": 3.9727,
+      "step": 1038336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2991251098198796e-05,
+      "loss": 3.9755,
+      "step": 1038848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.298286515068827e-05,
+      "loss": 3.9707,
+      "step": 1039360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.297447920317775e-05,
+      "loss": 3.9762,
+      "step": 1039872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.296609325566723e-05,
+      "loss": 3.9786,
+      "step": 1040384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.295770730815671e-05,
+      "loss": 3.9846,
+      "step": 1040896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.294932136064619e-05,
+      "loss": 3.9539,
+      "step": 1041408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.29409517919394e-05,
+      "loss": 3.9702,
+      "step": 1041920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.293256584442888e-05,
+      "loss": 3.968,
+      "step": 1042432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.292417989691836e-05,
+      "loss": 3.9775,
+      "step": 1042944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2915810328211574e-05,
+      "loss": 3.9835,
+      "step": 1043456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2907424380701054e-05,
+      "loss": 3.9718,
+      "step": 1043968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2899038433190534e-05,
+      "loss": 3.9687,
+      "step": 1044480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2890652485680014e-05,
+      "loss": 3.9742,
+      "step": 1044992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2882266538169494e-05,
+      "loss": 3.971,
+      "step": 1045504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.28738969694627e-05,
+      "loss": 3.9678,
+      "step": 1046016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.286551102195218e-05,
+      "loss": 3.9647,
+      "step": 1046528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.285712507444166e-05,
+      "loss": 3.9759,
+      "step": 1047040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.284873912693114e-05,
+      "loss": 3.9664,
+      "step": 1047552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.284035317942062e-05,
+      "loss": 3.9657,
+      "step": 1048064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.28319672319101e-05,
+      "loss": 3.9688,
+      "step": 1048576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.282358128439958e-05,
+      "loss": 3.9574,
+      "step": 1049088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.281519533688906e-05,
+      "loss": 3.976,
+      "step": 1049600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.280680938937854e-05,
+      "loss": 3.9681,
+      "step": 1050112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.279842344186803e-05,
+      "loss": 3.9701,
+      "step": 1050624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.279003749435751e-05,
+      "loss": 3.9748,
+      "step": 1051136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.278165154684698e-05,
+      "loss": 3.9739,
+      "step": 1051648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.27732819781402e-05,
+      "loss": 3.9726,
+      "step": 1052160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.276489603062968e-05,
+      "loss": 3.9674,
+      "step": 1052672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.275651008311916e-05,
+      "loss": 3.9793,
+      "step": 1053184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.274812413560863e-05,
+      "loss": 3.9687,
+      "step": 1053696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.273973818809811e-05,
+      "loss": 3.9628,
+      "step": 1054208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.273135224058759e-05,
+      "loss": 3.9766,
+      "step": 1054720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.272296629307707e-05,
+      "loss": 3.9714,
+      "step": 1055232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.271458034556655e-05,
+      "loss": 3.971,
+      "step": 1055744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.270621077685977e-05,
+      "loss": 3.9668,
+      "step": 1056256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.269782482934925e-05,
+      "loss": 3.9707,
+      "step": 1056768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.268943888183873e-05,
+      "loss": 3.9684,
+      "step": 1057280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.268105293432821e-05,
+      "loss": 3.9624,
+      "step": 1057792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.267266698681769e-05,
+      "loss": 3.964,
+      "step": 1058304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2664297418110896e-05,
+      "loss": 3.9783,
+      "step": 1058816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2655911470600376e-05,
+      "loss": 3.9829,
+      "step": 1059328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2647541901893585e-05,
+      "loss": 3.9637,
+      "step": 1059840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2639155954383065e-05,
+      "loss": 3.9679,
+      "step": 1060352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2630770006872545e-05,
+      "loss": 3.9831,
+      "step": 1060864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2622384059362025e-05,
+      "loss": 3.9715,
+      "step": 1061376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2613998111851505e-05,
+      "loss": 3.9645,
+      "step": 1061888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2605612164340985e-05,
+      "loss": 3.9673,
+      "step": 1062400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2597226216830465e-05,
+      "loss": 3.9827,
+      "step": 1062912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.258884026931995e-05,
+      "loss": 3.9798,
+      "step": 1063424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.258047070061316e-05,
+      "loss": 3.9685,
+      "step": 1063936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.257208475310264e-05,
+      "loss": 3.963,
+      "step": 1064448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.256369880559212e-05,
+      "loss": 3.9766,
+      "step": 1064960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.25553128580816e-05,
+      "loss": 3.9762,
+      "step": 1065472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.254695966817854e-05,
+      "loss": 3.9608,
+      "step": 1065984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.253857372066802e-05,
+      "loss": 3.9823,
+      "step": 1066496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.25301877731575e-05,
+      "loss": 3.9666,
+      "step": 1067008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.252180182564698e-05,
+      "loss": 3.9764,
+      "step": 1067520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.251341587813646e-05,
+      "loss": 3.9743,
+      "step": 1068032
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.080451011657715,
+      "eval_runtime": 291.8577,
+      "eval_samples_per_second": 1307.456,
+      "eval_steps_per_second": 40.859,
+      "step": 1068480
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.250502993062594e-05,
+      "loss": 3.9706,
+      "step": 1068544
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.249664398311542e-05,
+      "loss": 3.9627,
+      "step": 1069056
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.24882580356049e-05,
+      "loss": 3.9708,
+      "step": 1069568
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2479872088094385e-05,
+      "loss": 3.9798,
+      "step": 1070080
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2471502519387594e-05,
+      "loss": 3.9735,
+      "step": 1070592
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2463116571877074e-05,
+      "loss": 3.9722,
+      "step": 1071104
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.245474700317028e-05,
+      "loss": 3.9705,
+      "step": 1071616
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.244636105565976e-05,
+      "loss": 3.9676,
+      "step": 1072128
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.243797510814924e-05,
+      "loss": 3.9672,
+      "step": 1072640
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.242958916063872e-05,
+      "loss": 3.9741,
+      "step": 1073152
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.242121959193193e-05,
+      "loss": 3.9678,
+      "step": 1073664
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.241283364442141e-05,
+      "loss": 3.9776,
+      "step": 1074176
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.240446407571463e-05,
+      "loss": 3.9777,
+      "step": 1074688
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.239607812820411e-05,
+      "loss": 3.9566,
+      "step": 1075200
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.238769218069359e-05,
+      "loss": 3.9666,
+      "step": 1075712
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.237930623318307e-05,
+      "loss": 3.9552,
+      "step": 1076224
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.237092028567255e-05,
+      "loss": 3.96,
+      "step": 1076736
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.236253433816203e-05,
+      "loss": 3.9691,
+      "step": 1077248
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.235414839065151e-05,
+      "loss": 3.96,
+      "step": 1077760
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.234576244314099e-05,
+      "loss": 3.9734,
+      "step": 1078272
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.233737649563047e-05,
+      "loss": 3.9851,
+      "step": 1078784
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.232899054811994e-05,
+      "loss": 3.9637,
+      "step": 1079296
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.232060460060942e-05,
+      "loss": 3.9743,
+      "step": 1079808
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2312235031902637e-05,
+      "loss": 3.9757,
+      "step": 1080320
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.230384908439211e-05,
+      "loss": 3.958,
+      "step": 1080832
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.229546313688159e-05,
+      "loss": 3.9739,
+      "step": 1081344
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2287077189371076e-05,
+      "loss": 3.9617,
+      "step": 1081856
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2278691241860556e-05,
+      "loss": 3.9631,
+      "step": 1082368
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2270305294350036e-05,
+      "loss": 3.9647,
+      "step": 1082880
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2261919346839516e-05,
+      "loss": 3.9572,
+      "step": 1083392
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2253549778132725e-05,
+      "loss": 3.963,
+      "step": 1083904
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2245163830622205e-05,
+      "loss": 3.9707,
+      "step": 1084416
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2236777883111685e-05,
+      "loss": 3.975,
+      "step": 1084928
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2228391935601165e-05,
+      "loss": 3.9664,
+      "step": 1085440
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2220005988090645e-05,
+      "loss": 3.9601,
+      "step": 1085952
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2211636419383854e-05,
+      "loss": 3.9657,
+      "step": 1086464
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2203250471873334e-05,
+      "loss": 3.981,
+      "step": 1086976
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2194864524362814e-05,
+      "loss": 3.9588,
+      "step": 1087488
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2186478576852294e-05,
+      "loss": 3.9627,
+      "step": 1088000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.217810900814551e-05,
+      "loss": 3.9635,
+      "step": 1088512
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.216972306063499e-05,
+      "loss": 3.9603,
+      "step": 1089024
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.216133711312447e-05,
+      "loss": 3.9651,
+      "step": 1089536
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.215295116561395e-05,
+      "loss": 3.9718,
+      "step": 1090048
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.214456521810343e-05,
+      "loss": 3.9758,
+      "step": 1090560
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.213617927059291e-05,
+      "loss": 3.9693,
+      "step": 1091072
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.212780970188612e-05,
+      "loss": 3.9646,
+      "step": 1091584
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.21194237543756e-05,
+      "loss": 3.9657,
+      "step": 1092096
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.211103780686508e-05,
+      "loss": 3.9726,
+      "step": 1092608
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.210265185935456e-05,
+      "loss": 3.9489,
+      "step": 1093120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.209426591184404e-05,
+      "loss": 3.9692,
+      "step": 1093632
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.208589634313725e-05,
+      "loss": 3.9695,
+      "step": 1094144
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.207751039562673e-05,
+      "loss": 3.9729,
+      "step": 1094656
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2069124448116215e-05,
+      "loss": 3.9633,
+      "step": 1095168
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2060738500605694e-05,
+      "loss": 3.9575,
+      "step": 1095680
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2052368931898904e-05,
+      "loss": 3.9651,
+      "step": 1096192
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2043982984388384e-05,
+      "loss": 3.9565,
+      "step": 1096704
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2035597036877863e-05,
+      "loss": 3.9788,
+      "step": 1097216
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2027211089367343e-05,
+      "loss": 3.9642,
+      "step": 1097728
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.201884152066055e-05,
+      "loss": 3.981,
+      "step": 1098240
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.201045557315003e-05,
+      "loss": 3.9646,
+      "step": 1098752
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.200206962563951e-05,
+      "loss": 3.9488,
+      "step": 1099264
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.199368367812899e-05,
+      "loss": 3.9788,
+      "step": 1099776
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.198529773061847e-05,
+      "loss": 3.9557,
+      "step": 1100288
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.197691178310795e-05,
+      "loss": 3.9565,
+      "step": 1100800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.196852583559743e-05,
+      "loss": 3.9667,
+      "step": 1101312
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.196015626689065e-05,
+      "loss": 3.9708,
+      "step": 1101824
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.195177031938013e-05,
+      "loss": 3.9553,
+      "step": 1102336
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.19433843718696e-05,
+      "loss": 3.9581,
+      "step": 1102848
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.193499842435908e-05,
+      "loss": 3.9552,
+      "step": 1103360
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.192661247684856e-05,
+      "loss": 3.9525,
+      "step": 1103872
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.191822652933804e-05,
+      "loss": 3.9746,
+      "step": 1104384
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.190984058182752e-05,
+      "loss": 3.9674,
+      "step": 1104896
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1901454634317e-05,
+      "loss": 3.9621,
+      "step": 1105408
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.189306868680648e-05,
+      "loss": 3.961,
+      "step": 1105920
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.188468273929596e-05,
+      "loss": 3.9741,
+      "step": 1106432
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.187629679178544e-05,
+      "loss": 3.9639,
+      "step": 1106944
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.186791084427492e-05,
+      "loss": 3.9679,
+      "step": 1107456
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.185954127556813e-05,
+      "loss": 3.9621,
+      "step": 1107968
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.185115532805762e-05,
+      "loss": 3.9638,
+      "step": 1108480
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.18427693805471e-05,
+      "loss": 3.9676,
+      "step": 1108992
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1834399811840306e-05,
+      "loss": 3.9666,
+      "step": 1109504
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1826030243133515e-05,
+      "loss": 3.964,
+      "step": 1110016
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1817644295622995e-05,
+      "loss": 3.9723,
+      "step": 1110528
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1809258348112475e-05,
+      "loss": 3.9676,
+      "step": 1111040
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1800872400601955e-05,
+      "loss": 3.9499,
+      "step": 1111552
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1792486453091435e-05,
+      "loss": 3.9633,
+      "step": 1112064
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1784100505580915e-05,
+      "loss": 3.9723,
+      "step": 1112576
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1775714558070395e-05,
+      "loss": 3.9584,
+      "step": 1113088
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1767328610559875e-05,
+      "loss": 3.9623,
+      "step": 1113600
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1758942663049354e-05,
+      "loss": 3.9574,
+      "step": 1114112
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1750556715538834e-05,
+      "loss": 3.9605,
+      "step": 1114624
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.174218714683205e-05,
+      "loss": 3.961,
+      "step": 1115136
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.173380119932153e-05,
+      "loss": 3.9618,
+      "step": 1115648
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.172541525181101e-05,
+      "loss": 3.9653,
+      "step": 1116160
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.171702930430049e-05,
+      "loss": 3.9619,
+      "step": 1116672
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.17086597355937e-05,
+      "loss": 3.9756,
+      "step": 1117184
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.170027378808318e-05,
+      "loss": 3.9387,
+      "step": 1117696
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.169188784057266e-05,
+      "loss": 3.9613,
+      "step": 1118208
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.168350189306213e-05,
+      "loss": 3.9564,
+      "step": 1118720
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.167511594555161e-05,
+      "loss": 3.9661,
+      "step": 1119232
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.166672999804109e-05,
+      "loss": 3.9696,
+      "step": 1119744
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.165834405053057e-05,
+      "loss": 3.9622,
+      "step": 1120256
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.164995810302005e-05,
+      "loss": 3.9563,
+      "step": 1120768
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.164157215550954e-05,
+      "loss": 3.9587,
+      "step": 1121280
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.163318620799902e-05,
+      "loss": 3.9624,
+      "step": 1121792
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.16248002604885e-05,
+      "loss": 3.9543,
+      "step": 1122304
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.161643069178171e-05,
+      "loss": 3.9554,
+      "step": 1122816
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.160804474427119e-05,
+      "loss": 3.9605,
+      "step": 1123328
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.159965879676067e-05,
+      "loss": 3.9601,
+      "step": 1123840
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.159127284925015e-05,
+      "loss": 3.9492,
+      "step": 1124352
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.158288690173963e-05,
+      "loss": 3.9591,
+      "step": 1124864
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.157450095422911e-05,
+      "loss": 3.9455,
+      "step": 1125376
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.156611500671859e-05,
+      "loss": 3.9658,
+      "step": 1125888
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.155772905920807e-05,
+      "loss": 3.9519,
+      "step": 1126400
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.154934311169755e-05,
+      "loss": 3.9601,
+      "step": 1126912
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.154097354299076e-05,
+      "loss": 3.9625,
+      "step": 1127424
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.153258759548024e-05,
+      "loss": 3.9613,
+      "step": 1127936
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.152423440557718e-05,
+      "loss": 3.9573,
+      "step": 1128448
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.151584845806666e-05,
+      "loss": 3.9587,
+      "step": 1128960
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.150746251055614e-05,
+      "loss": 3.9649,
+      "step": 1129472
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.149907656304562e-05,
+      "loss": 3.9535,
+      "step": 1129984
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.14906906155351e-05,
+      "loss": 3.9504,
+      "step": 1130496
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.148230466802458e-05,
+      "loss": 3.9639,
+      "step": 1131008
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.147391872051406e-05,
+      "loss": 3.9626,
+      "step": 1131520
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.146553277300354e-05,
+      "loss": 3.9617,
+      "step": 1132032
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.145714682549302e-05,
+      "loss": 3.9533,
+      "step": 1132544
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1448760877982494e-05,
+      "loss": 3.9569,
+      "step": 1133056
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1440374930471974e-05,
+      "loss": 3.9563,
+      "step": 1133568
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.143200536176519e-05,
+      "loss": 3.9508,
+      "step": 1134080
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.142361941425467e-05,
+      "loss": 3.9514,
+      "step": 1134592
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.141523346674415e-05,
+      "loss": 3.9688,
+      "step": 1135104
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.140684751923363e-05,
+      "loss": 3.9706,
+      "step": 1135616
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.139846157172311e-05,
+      "loss": 3.9551,
+      "step": 1136128
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.139007562421259e-05,
+      "loss": 3.9561,
+      "step": 1136640
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.138168967670207e-05,
+      "loss": 3.9665,
+      "step": 1137152
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.137332010799528e-05,
+      "loss": 3.96,
+      "step": 1137664
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.136493416048476e-05,
+      "loss": 3.9542,
+      "step": 1138176
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.135654821297424e-05,
+      "loss": 3.9601,
+      "step": 1138688
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.134816226546372e-05,
+      "loss": 3.9692,
+      "step": 1139200
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.13397763179532e-05,
+      "loss": 3.9669,
+      "step": 1139712
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.133140674924641e-05,
+      "loss": 3.9568,
+      "step": 1140224
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1323020801735895e-05,
+      "loss": 3.951,
+      "step": 1140736
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1314634854225375e-05,
+      "loss": 3.9624,
+      "step": 1141248
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1306248906714855e-05,
+      "loss": 3.9649,
+      "step": 1141760
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1297879338008064e-05,
+      "loss": 3.9515,
+      "step": 1142272
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1289493390497544e-05,
+      "loss": 3.9724,
+      "step": 1142784
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1281107442987024e-05,
+      "loss": 3.9536,
+      "step": 1143296
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1272721495476504e-05,
+      "loss": 3.9647,
+      "step": 1143808
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.126435192676971e-05,
+      "loss": 3.9592,
+      "step": 1144320
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.073639869689941,
+      "eval_runtime": 293.6313,
+      "eval_samples_per_second": 1299.558,
+      "eval_steps_per_second": 40.612,
+      "step": 1144800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.125596597925919e-05,
+      "loss": 3.9566,
+      "step": 1144832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.124758003174867e-05,
+      "loss": 3.9546,
+      "step": 1145344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.123919408423815e-05,
+      "loss": 3.9607,
+      "step": 1145856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.123080813672763e-05,
+      "loss": 3.9663,
+      "step": 1146368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.122242218921711e-05,
+      "loss": 3.9619,
+      "step": 1146880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.121403624170659e-05,
+      "loss": 3.9622,
+      "step": 1147392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.120565029419608e-05,
+      "loss": 3.9548,
+      "step": 1147904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.119726434668556e-05,
+      "loss": 3.9562,
+      "step": 1148416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.118887839917504e-05,
+      "loss": 3.956,
+      "step": 1148928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.118049245166451e-05,
+      "loss": 3.9637,
+      "step": 1149440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.117210650415399e-05,
+      "loss": 3.9577,
+      "step": 1149952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.116372055664347e-05,
+      "loss": 3.9631,
+      "step": 1150464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.115535098793668e-05,
+      "loss": 3.9684,
+      "step": 1150976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.114696504042616e-05,
+      "loss": 3.9477,
+      "step": 1151488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.113857909291564e-05,
+      "loss": 3.9561,
+      "step": 1152000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.113019314540512e-05,
+      "loss": 3.9397,
+      "step": 1152512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.11218071978946e-05,
+      "loss": 3.9511,
+      "step": 1153024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.111342125038408e-05,
+      "loss": 3.9574,
+      "step": 1153536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.110503530287356e-05,
+      "loss": 3.9471,
+      "step": 1154048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.109664935536305e-05,
+      "loss": 3.9578,
+      "step": 1154560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.108826340785253e-05,
+      "loss": 3.9726,
+      "step": 1155072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.107987746034201e-05,
+      "loss": 3.9529,
+      "step": 1155584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.107150789163522e-05,
+      "loss": 3.962,
+      "step": 1156096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.10631219441247e-05,
+      "loss": 3.9662,
+      "step": 1156608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.105473599661418e-05,
+      "loss": 3.9504,
+      "step": 1157120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.104635004910366e-05,
+      "loss": 3.957,
+      "step": 1157632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.1037964101593137e-05,
+      "loss": 3.9453,
+      "step": 1158144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.1029578154082617e-05,
+      "loss": 3.9558,
+      "step": 1158656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.1021208585375826e-05,
+      "loss": 3.9549,
+      "step": 1159168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.1012822637865306e-05,
+      "loss": 3.9447,
+      "step": 1159680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1004436690354786e-05,
+      "loss": 3.9552,
+      "step": 1160192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0996050742844266e-05,
+      "loss": 3.9502,
+      "step": 1160704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0987664795333745e-05,
+      "loss": 3.9626,
+      "step": 1161216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0979278847823225e-05,
+      "loss": 3.9603,
+      "step": 1161728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.097090927911644e-05,
+      "loss": 3.9469,
+      "step": 1162240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.096253971040965e-05,
+      "loss": 3.954,
+      "step": 1162752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.095415376289913e-05,
+      "loss": 3.9682,
+      "step": 1163264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.094576781538861e-05,
+      "loss": 3.9478,
+      "step": 1163776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.093738186787809e-05,
+      "loss": 3.9517,
+      "step": 1164288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.092899592036757e-05,
+      "loss": 3.9499,
+      "step": 1164800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0920609972857043e-05,
+      "loss": 3.9497,
+      "step": 1165312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.091222402534652e-05,
+      "loss": 3.9504,
+      "step": 1165824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.090385445663974e-05,
+      "loss": 3.9685,
+      "step": 1166336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.089546850912922e-05,
+      "loss": 3.9601,
+      "step": 1166848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.08870825616187e-05,
+      "loss": 3.9556,
+      "step": 1167360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.087869661410818e-05,
+      "loss": 3.9581,
+      "step": 1167872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.087031066659766e-05,
+      "loss": 3.9489,
+      "step": 1168384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.086192471908714e-05,
+      "loss": 3.9644,
+      "step": 1168896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.085353877157662e-05,
+      "loss": 3.9363,
+      "step": 1169408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.084516920286983e-05,
+      "loss": 3.9568,
+      "step": 1169920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.083678325535931e-05,
+      "loss": 3.9604,
+      "step": 1170432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.082839730784879e-05,
+      "loss": 3.9615,
+      "step": 1170944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.082001136033827e-05,
+      "loss": 3.9533,
+      "step": 1171456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.081162541282775e-05,
+      "loss": 3.9441,
+      "step": 1171968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.080323946531723e-05,
+      "loss": 3.9531,
+      "step": 1172480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.079485351780671e-05,
+      "loss": 3.9425,
+      "step": 1172992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.078646757029619e-05,
+      "loss": 3.9702,
+      "step": 1173504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0778098001589404e-05,
+      "loss": 3.9496,
+      "step": 1174016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.076972843288261e-05,
+      "loss": 3.9687,
+      "step": 1174528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.076134248537209e-05,
+      "loss": 3.9512,
+      "step": 1175040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.075295653786157e-05,
+      "loss": 3.9449,
+      "step": 1175552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.074457059035105e-05,
+      "loss": 3.9613,
+      "step": 1176064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.073618464284053e-05,
+      "loss": 3.9438,
+      "step": 1176576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.072779869533001e-05,
+      "loss": 3.9521,
+      "step": 1177088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.071941274781949e-05,
+      "loss": 3.947,
+      "step": 1177600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.071102680030897e-05,
+      "loss": 3.9622,
+      "step": 1178112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.070265723160218e-05,
+      "loss": 3.9428,
+      "step": 1178624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.069427128409166e-05,
+      "loss": 3.9481,
+      "step": 1179136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.068588533658114e-05,
+      "loss": 3.9414,
+      "step": 1179648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.067749938907062e-05,
+      "loss": 3.9422,
+      "step": 1180160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.06691134415601e-05,
+      "loss": 3.9628,
+      "step": 1180672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.066074387285332e-05,
+      "loss": 3.9581,
+      "step": 1181184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.06523579253428e-05,
+      "loss": 3.9477,
+      "step": 1181696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.064397197783228e-05,
+      "loss": 3.9517,
+      "step": 1182208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.063558603032176e-05,
+      "loss": 3.9597,
+      "step": 1182720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0627216461614966e-05,
+      "loss": 3.9551,
+      "step": 1183232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0618863271711904e-05,
+      "loss": 3.9551,
+      "step": 1183744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0610477324201384e-05,
+      "loss": 3.9537,
+      "step": 1184256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0602091376690864e-05,
+      "loss": 3.9527,
+      "step": 1184768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0593705429180344e-05,
+      "loss": 3.956,
+      "step": 1185280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0585319481669824e-05,
+      "loss": 3.9526,
+      "step": 1185792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.057693353415931e-05,
+      "loss": 3.9532,
+      "step": 1186304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.056854758664879e-05,
+      "loss": 3.9625,
+      "step": 1186816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.056016163913827e-05,
+      "loss": 3.9545,
+      "step": 1187328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.055177569162775e-05,
+      "loss": 3.9373,
+      "step": 1187840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.054338974411723e-05,
+      "loss": 3.953,
+      "step": 1188352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0535003796606704e-05,
+      "loss": 3.957,
+      "step": 1188864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0526617849096184e-05,
+      "loss": 3.9493,
+      "step": 1189376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.05182482803894e-05,
+      "loss": 3.9526,
+      "step": 1189888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.050986233287888e-05,
+      "loss": 3.9443,
+      "step": 1190400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0501476385368356e-05,
+      "loss": 3.9518,
+      "step": 1190912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0493090437857836e-05,
+      "loss": 3.9476,
+      "step": 1191424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0484704490347316e-05,
+      "loss": 3.9473,
+      "step": 1191936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0476318542836796e-05,
+      "loss": 3.9529,
+      "step": 1192448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0467932595326276e-05,
+      "loss": 3.9493,
+      "step": 1192960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0459546647815756e-05,
+      "loss": 3.966,
+      "step": 1193472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.04511934579127e-05,
+      "loss": 3.9342,
+      "step": 1193984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0442807510402178e-05,
+      "loss": 3.943,
+      "step": 1194496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0434421562891658e-05,
+      "loss": 3.9507,
+      "step": 1195008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0426035615381138e-05,
+      "loss": 3.9499,
+      "step": 1195520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0417649667870618e-05,
+      "loss": 3.9592,
+      "step": 1196032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0409263720360097e-05,
+      "loss": 3.9531,
+      "step": 1196544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0400877772849577e-05,
+      "loss": 3.9442,
+      "step": 1197056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0392491825339057e-05,
+      "loss": 3.9487,
+      "step": 1197568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.038412225663227e-05,
+      "loss": 3.9508,
+      "step": 1198080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.037573630912175e-05,
+      "loss": 3.9408,
+      "step": 1198592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.036736674041496e-05,
+      "loss": 3.9402,
+      "step": 1199104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.035898079290444e-05,
+      "loss": 3.9512,
+      "step": 1199616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.035059484539392e-05,
+      "loss": 3.9504,
+      "step": 1200128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0342208897883402e-05,
+      "loss": 3.9389,
+      "step": 1200640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0333822950372882e-05,
+      "loss": 3.9442,
+      "step": 1201152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0325437002862362e-05,
+      "loss": 3.9359,
+      "step": 1201664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.031706743415557e-05,
+      "loss": 3.9537,
+      "step": 1202176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.030868148664505e-05,
+      "loss": 3.9434,
+      "step": 1202688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.030029553913453e-05,
+      "loss": 3.9459,
+      "step": 1203200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.029190959162401e-05,
+      "loss": 3.951,
+      "step": 1203712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0283540022917224e-05,
+      "loss": 3.9506,
+      "step": 1204224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0275170454210433e-05,
+      "loss": 3.948,
+      "step": 1204736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0266784506699913e-05,
+      "loss": 3.9469,
+      "step": 1205248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0258398559189393e-05,
+      "loss": 3.9533,
+      "step": 1205760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0250012611678873e-05,
+      "loss": 3.945,
+      "step": 1206272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0241626664168356e-05,
+      "loss": 3.9398,
+      "step": 1206784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0233240716657836e-05,
+      "loss": 3.953,
+      "step": 1207296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0224854769147316e-05,
+      "loss": 3.9457,
+      "step": 1207808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0216468821636796e-05,
+      "loss": 3.9553,
+      "step": 1208320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0208082874126276e-05,
+      "loss": 3.9413,
+      "step": 1208832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0199696926615756e-05,
+      "loss": 3.9468,
+      "step": 1209344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0191310979105236e-05,
+      "loss": 3.9434,
+      "step": 1209856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0182925031594716e-05,
+      "loss": 3.9432,
+      "step": 1210368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0174539084084192e-05,
+      "loss": 3.9347,
+      "step": 1210880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0166169515377408e-05,
+      "loss": 3.9559,
+      "step": 1211392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0157799946670617e-05,
+      "loss": 3.9606,
+      "step": 1211904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0149413999160097e-05,
+      "loss": 3.9472,
+      "step": 1212416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.014104443045331e-05,
+      "loss": 3.942,
+      "step": 1212928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.013265848294279e-05,
+      "loss": 3.955,
+      "step": 1213440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.012427253543227e-05,
+      "loss": 3.9507,
+      "step": 1213952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.011588658792175e-05,
+      "loss": 3.9449,
+      "step": 1214464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.010750064041123e-05,
+      "loss": 3.9492,
+      "step": 1214976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.009911469290071e-05,
+      "loss": 3.9563,
+      "step": 1215488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.009072874539019e-05,
+      "loss": 3.9525,
+      "step": 1216000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0082342797879666e-05,
+      "loss": 3.948,
+      "step": 1216512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0073973229172882e-05,
+      "loss": 3.9408,
+      "step": 1217024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.006558728166236e-05,
+      "loss": 3.9535,
+      "step": 1217536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0057201334151835e-05,
+      "loss": 3.9515,
+      "step": 1218048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0048815386641315e-05,
+      "loss": 3.9383,
+      "step": 1218560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.004044581793453e-05,
+      "loss": 3.9621,
+      "step": 1219072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.003205987042401e-05,
+      "loss": 3.9437,
+      "step": 1219584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0023673922913487e-05,
+      "loss": 3.9532,
+      "step": 1220096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0015287975402967e-05,
+      "loss": 3.953,
+      "step": 1220608
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 3.0006902027892447e-05,
+      "loss": 3.9441,
+      "step": 1221120
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.068778038024902,
+      "eval_runtime": 281.9777,
+      "eval_samples_per_second": 1353.266,
+      "eval_steps_per_second": 42.291,
+      "step": 1221120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9998516080381927e-05,
+      "loss": 3.9429,
+      "step": 1221632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9990130132871407e-05,
+      "loss": 3.9463,
+      "step": 1222144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9981744185360887e-05,
+      "loss": 3.9571,
+      "step": 1222656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.99733746166541e-05,
+      "loss": 3.9485,
+      "step": 1223168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.996498866914358e-05,
+      "loss": 3.9527,
+      "step": 1223680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.995660272163306e-05,
+      "loss": 3.9438,
+      "step": 1224192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.994821677412254e-05,
+      "loss": 3.9468,
+      "step": 1224704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.993983082661202e-05,
+      "loss": 3.9448,
+      "step": 1225216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.99314448791015e-05,
+      "loss": 3.9529,
+      "step": 1225728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.992305893159098e-05,
+      "loss": 3.9404,
+      "step": 1226240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.991467298408046e-05,
+      "loss": 3.957,
+      "step": 1226752
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.99063197941774e-05,
+      "loss": 3.9585,
+      "step": 1227264
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.989793384666688e-05,
+      "loss": 3.9379,
+      "step": 1227776
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.988954789915636e-05,
+      "loss": 3.9436,
+      "step": 1228288
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.988116195164584e-05,
+      "loss": 3.9332,
+      "step": 1228800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.987277600413532e-05,
+      "loss": 3.9386,
+      "step": 1229312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9864390056624804e-05,
+      "loss": 3.947,
+      "step": 1229824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9856004109114284e-05,
+      "loss": 3.9391,
+      "step": 1230336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9847618161603764e-05,
+      "loss": 3.9445,
+      "step": 1230848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9839232214093244e-05,
+      "loss": 3.9658,
+      "step": 1231360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9830846266582724e-05,
+      "loss": 3.9416,
+      "step": 1231872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9822460319072197e-05,
+      "loss": 3.9498,
+      "step": 1232384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.981407437156168e-05,
+      "loss": 3.951,
+      "step": 1232896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9805704802854896e-05,
+      "loss": 3.9447,
+      "step": 1233408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9797335234148105e-05,
+      "loss": 3.947,
+      "step": 1233920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9788949286637585e-05,
+      "loss": 3.9347,
+      "step": 1234432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9780563339127065e-05,
+      "loss": 3.9438,
+      "step": 1234944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9772177391616545e-05,
+      "loss": 3.9466,
+      "step": 1235456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9763791444106025e-05,
+      "loss": 3.9329,
+      "step": 1235968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.97554054965955e-05,
+      "loss": 3.9442,
+      "step": 1236480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.974701954908498e-05,
+      "loss": 3.9402,
+      "step": 1236992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.973863360157446e-05,
+      "loss": 3.9519,
+      "step": 1237504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.973026403286767e-05,
+      "loss": 3.9469,
+      "step": 1238016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.972187808535715e-05,
+      "loss": 3.9381,
+      "step": 1238528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9713508516650366e-05,
+      "loss": 3.9453,
+      "step": 1239040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.970512256913985e-05,
+      "loss": 3.9577,
+      "step": 1239552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9696736621629323e-05,
+      "loss": 3.9353,
+      "step": 1240064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9688350674118803e-05,
+      "loss": 3.9414,
+      "step": 1240576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9679964726608283e-05,
+      "loss": 3.9426,
+      "step": 1241088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9671578779097763e-05,
+      "loss": 3.935,
+      "step": 1241600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9663192831587243e-05,
+      "loss": 3.9421,
+      "step": 1242112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9654806884076726e-05,
+      "loss": 3.9568,
+      "step": 1242624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9646437315369935e-05,
+      "loss": 3.9504,
+      "step": 1243136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9638051367859415e-05,
+      "loss": 3.9426,
+      "step": 1243648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9629665420348895e-05,
+      "loss": 3.9486,
+      "step": 1244160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9621279472838375e-05,
+      "loss": 3.9388,
+      "step": 1244672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9612909904131588e-05,
+      "loss": 3.9514,
+      "step": 1245184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9604523956621068e-05,
+      "loss": 3.9276,
+      "step": 1245696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9596138009110548e-05,
+      "loss": 3.9428,
+      "step": 1246208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9587752061600027e-05,
+      "loss": 3.9517,
+      "step": 1246720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9579382492893237e-05,
+      "loss": 3.9484,
+      "step": 1247232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9570996545382717e-05,
+      "loss": 3.9451,
+      "step": 1247744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9562610597872196e-05,
+      "loss": 3.9324,
+      "step": 1248256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.955422465036168e-05,
+      "loss": 3.9442,
+      "step": 1248768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.954585508165489e-05,
+      "loss": 3.9301,
+      "step": 1249280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9537485512948098e-05,
+      "loss": 3.9571,
+      "step": 1249792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9529099565437578e-05,
+      "loss": 3.9393,
+      "step": 1250304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9520713617927058e-05,
+      "loss": 3.9544,
+      "step": 1250816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.951232767041654e-05,
+      "loss": 3.9424,
+      "step": 1251328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.950394172290602e-05,
+      "loss": 3.9364,
+      "step": 1251840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.94955557753955e-05,
+      "loss": 3.9451,
+      "step": 1252352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.948716982788498e-05,
+      "loss": 3.9369,
+      "step": 1252864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.947878388037446e-05,
+      "loss": 3.9412,
+      "step": 1253376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9470430690471403e-05,
+      "loss": 3.9353,
+      "step": 1253888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9462044742960883e-05,
+      "loss": 3.9508,
+      "step": 1254400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9453658795450363e-05,
+      "loss": 3.9346,
+      "step": 1254912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9445272847939843e-05,
+      "loss": 3.9344,
+      "step": 1255424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.943690327923305e-05,
+      "loss": 3.9301,
+      "step": 1255936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.942851733172253e-05,
+      "loss": 3.9317,
+      "step": 1256448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.942013138421201e-05,
+      "loss": 3.9537,
+      "step": 1256960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9411745436701495e-05,
+      "loss": 3.9481,
+      "step": 1257472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9403359489190975e-05,
+      "loss": 3.9388,
+      "step": 1257984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9394973541680455e-05,
+      "loss": 3.9418,
+      "step": 1258496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9386587594169935e-05,
+      "loss": 3.9484,
+      "step": 1259008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9378201646659415e-05,
+      "loss": 3.9463,
+      "step": 1259520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9369832077952624e-05,
+      "loss": 3.9395,
+      "step": 1260032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9361462509245836e-05,
+      "loss": 3.9455,
+      "step": 1260544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9353076561735316e-05,
+      "loss": 3.9419,
+      "step": 1261056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9344690614224796e-05,
+      "loss": 3.9454,
+      "step": 1261568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9336304666714276e-05,
+      "loss": 3.9405,
+      "step": 1262080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9327935098007485e-05,
+      "loss": 3.9441,
+      "step": 1262592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9319549150496965e-05,
+      "loss": 3.9526,
+      "step": 1263104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9311163202986445e-05,
+      "loss": 3.9439,
+      "step": 1263616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9302793634279658e-05,
+      "loss": 3.9274,
+      "step": 1264128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9294407686769138e-05,
+      "loss": 3.9418,
+      "step": 1264640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9286021739258618e-05,
+      "loss": 3.9433,
+      "step": 1265152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9277635791748098e-05,
+      "loss": 3.9447,
+      "step": 1265664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9269249844237578e-05,
+      "loss": 3.9401,
+      "step": 1266176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9260863896727057e-05,
+      "loss": 3.9353,
+      "step": 1266688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9252477949216537e-05,
+      "loss": 3.9438,
+      "step": 1267200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.924409200170602e-05,
+      "loss": 3.9348,
+      "step": 1267712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.923572243299923e-05,
+      "loss": 3.941,
+      "step": 1268224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.922733648548871e-05,
+      "loss": 3.9395,
+      "step": 1268736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.921895053797819e-05,
+      "loss": 3.9356,
+      "step": 1269248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.921056459046767e-05,
+      "loss": 3.9582,
+      "step": 1269760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.920217864295715e-05,
+      "loss": 3.9281,
+      "step": 1270272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.919379269544663e-05,
+      "loss": 3.9274,
+      "step": 1270784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9185406747936106e-05,
+      "loss": 3.9426,
+      "step": 1271296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9177020800425586e-05,
+      "loss": 3.9421,
+      "step": 1271808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9168651231718802e-05,
+      "loss": 3.9478,
+      "step": 1272320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9160265284208275e-05,
+      "loss": 3.939,
+      "step": 1272832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.915189571550149e-05,
+      "loss": 3.9338,
+      "step": 1273344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9143509767990974e-05,
+      "loss": 3.9363,
+      "step": 1273856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9135123820480454e-05,
+      "loss": 3.945,
+      "step": 1274368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9126737872969928e-05,
+      "loss": 3.9309,
+      "step": 1274880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9118368304263143e-05,
+      "loss": 3.9295,
+      "step": 1275392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9109982356752623e-05,
+      "loss": 3.9446,
+      "step": 1275904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9101596409242103e-05,
+      "loss": 3.9393,
+      "step": 1276416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.909321046173158e-05,
+      "loss": 3.9291,
+      "step": 1276928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.908482451422106e-05,
+      "loss": 3.9298,
+      "step": 1277440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.907643856671054e-05,
+      "loss": 3.9293,
+      "step": 1277952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.906806899800375e-05,
+      "loss": 3.9413,
+      "step": 1278464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.905968305049323e-05,
+      "loss": 3.9317,
+      "step": 1278976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9051297102982712e-05,
+      "loss": 3.9376,
+      "step": 1279488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9042911155472192e-05,
+      "loss": 3.9394,
+      "step": 1280000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9034525207961672e-05,
+      "loss": 3.9431,
+      "step": 1280512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.902615563925488e-05,
+      "loss": 3.9328,
+      "step": 1281024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.901776969174436e-05,
+      "loss": 3.9378,
+      "step": 1281536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.900938374423384e-05,
+      "loss": 3.9408,
+      "step": 1282048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.900099779672332e-05,
+      "loss": 3.9373,
+      "step": 1282560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8992611849212804e-05,
+      "loss": 3.9244,
+      "step": 1283072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8984242280506014e-05,
+      "loss": 3.9417,
+      "step": 1283584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8975856332995494e-05,
+      "loss": 3.9366,
+      "step": 1284096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8967470385484973e-05,
+      "loss": 3.9428,
+      "step": 1284608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8959084437974453e-05,
+      "loss": 3.9288,
+      "step": 1285120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8950698490463933e-05,
+      "loss": 3.9414,
+      "step": 1285632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8942312542953413e-05,
+      "loss": 3.9299,
+      "step": 1286144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8933942974246626e-05,
+      "loss": 3.9351,
+      "step": 1286656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8925557026736106e-05,
+      "loss": 3.9206,
+      "step": 1287168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8917171079225586e-05,
+      "loss": 3.9498,
+      "step": 1287680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8908785131715066e-05,
+      "loss": 3.9491,
+      "step": 1288192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8900399184204546e-05,
+      "loss": 3.9377,
+      "step": 1288704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8892013236694026e-05,
+      "loss": 3.9283,
+      "step": 1289216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8883627289183506e-05,
+      "loss": 3.9491,
+      "step": 1289728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.887524134167299e-05,
+      "loss": 3.9358,
+      "step": 1290240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8866871772966198e-05,
+      "loss": 3.9357,
+      "step": 1290752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8858485825455678e-05,
+      "loss": 3.9361,
+      "step": 1291264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8850099877945158e-05,
+      "loss": 3.9463,
+      "step": 1291776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8841713930434638e-05,
+      "loss": 3.9388,
+      "step": 1292288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.883334436172785e-05,
+      "loss": 3.9407,
+      "step": 1292800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.882497479302106e-05,
+      "loss": 3.9281,
+      "step": 1293312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.881658884551054e-05,
+      "loss": 3.945,
+      "step": 1293824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.880820289800002e-05,
+      "loss": 3.9394,
+      "step": 1294336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.87998169504895e-05,
+      "loss": 3.9305,
+      "step": 1294848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8791447381782712e-05,
+      "loss": 3.9516,
+      "step": 1295360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8783061434272192e-05,
+      "loss": 3.932,
+      "step": 1295872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8774675486761672e-05,
+      "loss": 3.9402,
+      "step": 1296384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.876628953925115e-05,
+      "loss": 3.9413,
+      "step": 1296896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.875791997054436e-05,
+      "loss": 3.9355,
+      "step": 1297408
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.0630364418029785,
+      "eval_runtime": 296.8585,
+      "eval_samples_per_second": 1285.431,
+      "eval_steps_per_second": 40.171,
+      "step": 1297440
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.874953402303384e-05,
+      "loss": 3.9302,
+      "step": 1297920
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.874114807552332e-05,
+      "loss": 3.9395,
+      "step": 1298432
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.87327621280128e-05,
+      "loss": 3.9452,
+      "step": 1298944
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8724376180502284e-05,
+      "loss": 3.9402,
+      "step": 1299456
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8715990232991764e-05,
+      "loss": 3.9451,
+      "step": 1299968
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8707604285481237e-05,
+      "loss": 3.9312,
+      "step": 1300480
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8699218337970717e-05,
+      "loss": 3.9341,
+      "step": 1300992
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8690848769263933e-05,
+      "loss": 3.9323,
+      "step": 1301504
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.868246282175341e-05,
+      "loss": 3.9449,
+      "step": 1302016
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8674093253046625e-05,
+      "loss": 3.9327,
+      "step": 1302528
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8665707305536105e-05,
+      "loss": 3.9428,
+      "step": 1303040
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8657321358025585e-05,
+      "loss": 3.9451,
+      "step": 1303552
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8648951789318794e-05,
+      "loss": 3.9298,
+      "step": 1304064
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8640565841808274e-05,
+      "loss": 3.931,
+      "step": 1304576
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8632179894297754e-05,
+      "loss": 3.9221,
+      "step": 1305088
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8623793946787238e-05,
+      "loss": 3.9326,
+      "step": 1305600
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.861540799927671e-05,
+      "loss": 3.9327,
+      "step": 1306112
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.860702205176619e-05,
+      "loss": 3.9283,
+      "step": 1306624
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.859863610425567e-05,
+      "loss": 3.9316,
+      "step": 1307136
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.859025015674515e-05,
+      "loss": 3.9583,
+      "step": 1307648
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.858186420923463e-05,
+      "loss": 3.9325,
+      "step": 1308160
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8573478261724114e-05,
+      "loss": 3.9372,
+      "step": 1308672
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8565092314213594e-05,
+      "loss": 3.9448,
+      "step": 1309184
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8556722745506803e-05,
+      "loss": 3.9343,
+      "step": 1309696
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8548336797996283e-05,
+      "loss": 3.9299,
+      "step": 1310208
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8539950850485763e-05,
+      "loss": 3.9258,
+      "step": 1310720
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8531564902975243e-05,
+      "loss": 3.9338,
+      "step": 1311232
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8523178955464723e-05,
+      "loss": 3.9339,
+      "step": 1311744
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8514809386757935e-05,
+      "loss": 3.9202,
+      "step": 1312256
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8506423439247415e-05,
+      "loss": 3.9352,
+      "step": 1312768
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8498037491736895e-05,
+      "loss": 3.928,
+      "step": 1313280
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8489651544226375e-05,
+      "loss": 3.9417,
+      "step": 1313792
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8481265596715855e-05,
+      "loss": 3.9409,
+      "step": 1314304
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8472879649205335e-05,
+      "loss": 3.9239,
+      "step": 1314816
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8464510080498548e-05,
+      "loss": 3.9398,
+      "step": 1315328
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8456140511791757e-05,
+      "loss": 3.9452,
+      "step": 1315840
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8447754564281237e-05,
+      "loss": 3.9236,
+      "step": 1316352
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8439368616770717e-05,
+      "loss": 3.9348,
+      "step": 1316864
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8430982669260197e-05,
+      "loss": 3.9299,
+      "step": 1317376
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8422596721749677e-05,
+      "loss": 3.9245,
+      "step": 1317888
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.841421077423916e-05,
+      "loss": 3.9333,
+      "step": 1318400
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.840584120553237e-05,
+      "loss": 3.9451,
+      "step": 1318912
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.839745525802185e-05,
+      "loss": 3.9394,
+      "step": 1319424
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.838906931051133e-05,
+      "loss": 3.934,
+      "step": 1319936
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.838068336300081e-05,
+      "loss": 3.9418,
+      "step": 1320448
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.837229741549029e-05,
+      "loss": 3.9292,
+      "step": 1320960
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.836391146797977e-05,
+      "loss": 3.9392,
+      "step": 1321472
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8355525520469245e-05,
+      "loss": 3.9194,
+      "step": 1321984
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.834715595176246e-05,
+      "loss": 3.9305,
+      "step": 1322496
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.833877000425194e-05,
+      "loss": 3.9417,
+      "step": 1323008
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.833038405674142e-05,
+      "loss": 3.9387,
+      "step": 1323520
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8321998109230898e-05,
+      "loss": 3.9321,
+      "step": 1324032
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8313612161720378e-05,
+      "loss": 3.9241,
+      "step": 1324544
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8305226214209858e-05,
+      "loss": 3.9373,
+      "step": 1325056
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8296840266699338e-05,
+      "loss": 3.9188,
+      "step": 1325568
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8288470697992547e-05,
+      "loss": 3.9458,
+      "step": 1326080
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8280084750482027e-05,
+      "loss": 3.9318,
+      "step": 1326592
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8271698802971507e-05,
+      "loss": 3.946,
+      "step": 1327104
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.826331285546099e-05,
+      "loss": 3.9324,
+      "step": 1327616
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.825492690795047e-05,
+      "loss": 3.9274,
+      "step": 1328128
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.824654096043995e-05,
+      "loss": 3.9307,
+      "step": 1328640
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.823815501292943e-05,
+      "loss": 3.9369,
+      "step": 1329152
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.822976906541891e-05,
+      "loss": 3.9255,
+      "step": 1329664
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.822138311790839e-05,
+      "loss": 3.9252,
+      "step": 1330176
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.821299717039787e-05,
+      "loss": 3.942,
+      "step": 1330688
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.820461122288735e-05,
+      "loss": 3.9291,
+      "step": 1331200
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8196241654180562e-05,
+      "loss": 3.9209,
+      "step": 1331712
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8187855706670042e-05,
+      "loss": 3.9236,
+      "step": 1332224
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.817948613796325e-05,
+      "loss": 3.9224,
+      "step": 1332736
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.817110019045273e-05,
+      "loss": 3.9408,
+      "step": 1333248
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.816271424294221e-05,
+      "loss": 3.9393,
+      "step": 1333760
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.815432829543169e-05,
+      "loss": 3.9286,
+      "step": 1334272
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.814594234792117e-05,
+      "loss": 3.9325,
+      "step": 1334784
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8137556400410654e-05,
+      "loss": 3.937,
+      "step": 1335296
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8129170452900134e-05,
+      "loss": 3.9357,
+      "step": 1335808
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8120800884193343e-05,
+      "loss": 3.9299,
+      "step": 1336320
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8112414936682823e-05,
+      "loss": 3.9365,
+      "step": 1336832
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8104028989172303e-05,
+      "loss": 3.9319,
+      "step": 1337344
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8095643041661783e-05,
+      "loss": 3.9406,
+      "step": 1337856
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.808725709415126e-05,
+      "loss": 3.9299,
+      "step": 1338368
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.807887114664074e-05,
+      "loss": 3.9358,
+      "step": 1338880
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.807048519913022e-05,
+      "loss": 3.9366,
+      "step": 1339392
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.80620992516197e-05,
+      "loss": 3.9363,
+      "step": 1339904
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.805372968291291e-05,
+      "loss": 3.9179,
+      "step": 1340416
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8045343735402392e-05,
+      "loss": 3.9318,
+      "step": 1340928
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8036957787891872e-05,
+      "loss": 3.9307,
+      "step": 1341440
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8028571840381352e-05,
+      "loss": 3.9352,
+      "step": 1341952
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.802020227167456e-05,
+      "loss": 3.9326,
+      "step": 1342464
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8011832702967777e-05,
+      "loss": 3.9215,
+      "step": 1342976
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8003446755457257e-05,
+      "loss": 3.9322,
+      "step": 1343488
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7995060807946733e-05,
+      "loss": 3.9246,
+      "step": 1344000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7986674860436213e-05,
+      "loss": 3.9348,
+      "step": 1344512
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7978288912925693e-05,
+      "loss": 3.9233,
+      "step": 1345024
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7969902965415173e-05,
+      "loss": 3.9297,
+      "step": 1345536
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7961517017904653e-05,
+      "loss": 3.9413,
+      "step": 1346048
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7953131070394133e-05,
+      "loss": 3.9308,
+      "step": 1346560
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7944761501687346e-05,
+      "loss": 3.9136,
+      "step": 1347072
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7936375554176826e-05,
+      "loss": 3.9316,
+      "step": 1347584
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7927989606666306e-05,
+      "loss": 3.9323,
+      "step": 1348096
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7919603659155786e-05,
+      "loss": 3.9364,
+      "step": 1348608
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7911234090448995e-05,
+      "loss": 3.9297,
+      "step": 1349120
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7902848142938475e-05,
+      "loss": 3.9263,
+      "step": 1349632
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7894478574231687e-05,
+      "loss": 3.9235,
+      "step": 1350144
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7886092626721167e-05,
+      "loss": 3.9361,
+      "step": 1350656
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7877706679210647e-05,
+      "loss": 3.9156,
+      "step": 1351168
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7869320731700127e-05,
+      "loss": 3.924,
+      "step": 1351680
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7860934784189607e-05,
+      "loss": 3.9322,
+      "step": 1352192
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7852548836679087e-05,
+      "loss": 3.9324,
+      "step": 1352704
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7844162889168567e-05,
+      "loss": 3.9122,
+      "step": 1353216
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7835776941658047e-05,
+      "loss": 3.9276,
+      "step": 1353728
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.782740737295126e-05,
+      "loss": 3.9162,
+      "step": 1354240
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.781902142544074e-05,
+      "loss": 3.9329,
+      "step": 1354752
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.781063547793022e-05,
+      "loss": 3.9236,
+      "step": 1355264
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.780226590922343e-05,
+      "loss": 3.9188,
+      "step": 1355776
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7793879961712908e-05,
+      "loss": 3.9376,
+      "step": 1356288
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.778549401420239e-05,
+      "loss": 3.9316,
+      "step": 1356800
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.77771244454956e-05,
+      "loss": 3.9228,
+      "step": 1357312
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.776873849798508e-05,
+      "loss": 3.9305,
+      "step": 1357824
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.776035255047456e-05,
+      "loss": 3.9283,
+      "step": 1358336
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.775196660296404e-05,
+      "loss": 3.931,
+      "step": 1358848
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.774358065545352e-05,
+      "loss": 3.9124,
+      "step": 1359360
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7735194707943e-05,
+      "loss": 3.9373,
+      "step": 1359872
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7726808760432484e-05,
+      "loss": 3.9236,
+      "step": 1360384
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7718422812921964e-05,
+      "loss": 3.9322,
+      "step": 1360896
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7710036865411444e-05,
+      "loss": 3.921,
+      "step": 1361408
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7701650917900917e-05,
+      "loss": 3.9291,
+      "step": 1361920
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7693264970390397e-05,
+      "loss": 3.9237,
+      "step": 1362432
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7684879022879877e-05,
+      "loss": 3.9255,
+      "step": 1362944
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.767649307536936e-05,
+      "loss": 3.9098,
+      "step": 1363456
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.766812350666257e-05,
+      "loss": 3.9424,
+      "step": 1363968
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.765973755915205e-05,
+      "loss": 3.935,
+      "step": 1364480
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.765135161164153e-05,
+      "loss": 3.9313,
+      "step": 1364992
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7642982042934738e-05,
+      "loss": 3.9208,
+      "step": 1365504
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.763459609542422e-05,
+      "loss": 3.9353,
+      "step": 1366016
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.76262101479137e-05,
+      "loss": 3.932,
+      "step": 1366528
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.761782420040318e-05,
+      "loss": 3.9255,
+      "step": 1367040
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.760943825289266e-05,
+      "loss": 3.9245,
+      "step": 1367552
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.760105230538214e-05,
+      "loss": 3.9339,
+      "step": 1368064
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.759266635787162e-05,
+      "loss": 3.9287,
+      "step": 1368576
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.75842804103611e-05,
+      "loss": 3.9381,
+      "step": 1369088
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7575910841654314e-05,
+      "loss": 3.9163,
+      "step": 1369600
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7567524894143794e-05,
+      "loss": 3.9338,
+      "step": 1370112
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7559138946633274e-05,
+      "loss": 3.9315,
+      "step": 1370624
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7550752999122754e-05,
+      "loss": 3.9231,
+      "step": 1371136
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7542383430415963e-05,
+      "loss": 3.94,
+      "step": 1371648
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7533997482905443e-05,
+      "loss": 3.9227,
+      "step": 1372160
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7525611535394923e-05,
+      "loss": 3.9297,
+      "step": 1372672
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7517225587884403e-05,
+      "loss": 3.9284,
+      "step": 1373184
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7508839640373886e-05,
+      "loss": 3.9289,
+      "step": 1373696
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.0588698387146,
+      "eval_runtime": 298.465,
+      "eval_samples_per_second": 1278.512,
+      "eval_steps_per_second": 39.954,
+      "step": 1373760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7500453692863366e-05,
+      "loss": 3.9194,
+      "step": 1374208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7492067745352846e-05,
+      "loss": 3.9312,
+      "step": 1374720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7483681797842326e-05,
+      "loss": 3.9341,
+      "step": 1375232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7475295850331806e-05,
+      "loss": 3.9312,
+      "step": 1375744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.746690990282128e-05,
+      "loss": 3.936,
+      "step": 1376256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7458523955310762e-05,
+      "loss": 3.9201,
+      "step": 1376768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7450138007800242e-05,
+      "loss": 3.9247,
+      "step": 1377280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7441752060289722e-05,
+      "loss": 3.9235,
+      "step": 1377792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7433366112779202e-05,
+      "loss": 3.9347,
+      "step": 1378304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7424980165268682e-05,
+      "loss": 3.9247,
+      "step": 1378816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7416594217758162e-05,
+      "loss": 3.9343,
+      "step": 1379328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7408208270247642e-05,
+      "loss": 3.9355,
+      "step": 1379840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7399838701540855e-05,
+      "loss": 3.9195,
+      "step": 1380352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7391452754030335e-05,
+      "loss": 3.9244,
+      "step": 1380864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7383066806519814e-05,
+      "loss": 3.9115,
+      "step": 1381376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7374680859009294e-05,
+      "loss": 3.9215,
+      "step": 1381888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7366294911498774e-05,
+      "loss": 3.9259,
+      "step": 1382400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7357908963988254e-05,
+      "loss": 3.9207,
+      "step": 1382912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7349523016477734e-05,
+      "loss": 3.9212,
+      "step": 1383424
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7341137068967214e-05,
+      "loss": 3.9466,
+      "step": 1383936
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.733275112145669e-05,
+      "loss": 3.9234,
+      "step": 1384448
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.732436517394617e-05,
+      "loss": 3.926,
+      "step": 1384960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.731597922643565e-05,
+      "loss": 3.9337,
+      "step": 1385472
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7307609657728867e-05,
+      "loss": 3.9301,
+      "step": 1385984
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.729922371021834e-05,
+      "loss": 3.9206,
+      "step": 1386496
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7290837762707823e-05,
+      "loss": 3.9161,
+      "step": 1387008
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.728246819400104e-05,
+      "loss": 3.9263,
+      "step": 1387520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.727408224649052e-05,
+      "loss": 3.9211,
+      "step": 1388032
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7265696298979992e-05,
+      "loss": 3.9146,
+      "step": 1388544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7257310351469472e-05,
+      "loss": 3.9267,
+      "step": 1389056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7248924403958952e-05,
+      "loss": 3.9189,
+      "step": 1389568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7240538456448432e-05,
+      "loss": 3.9305,
+      "step": 1390080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7232168887741645e-05,
+      "loss": 3.9279,
+      "step": 1390592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7223782940231124e-05,
+      "loss": 3.9169,
+      "step": 1391104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.721541337152434e-05,
+      "loss": 3.9293,
+      "step": 1391616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7207027424013814e-05,
+      "loss": 3.9384,
+      "step": 1392128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7198641476503293e-05,
+      "loss": 3.9123,
+      "step": 1392640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7190255528992777e-05,
+      "loss": 3.9271,
+      "step": 1393152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7181869581482257e-05,
+      "loss": 3.9186,
+      "step": 1393664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7173483633971737e-05,
+      "loss": 3.9194,
+      "step": 1394176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7165097686461217e-05,
+      "loss": 3.9208,
+      "step": 1394688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7156711738950697e-05,
+      "loss": 3.9378,
+      "step": 1395200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7148325791440177e-05,
+      "loss": 3.9277,
+      "step": 1395712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7139939843929657e-05,
+      "loss": 3.9242,
+      "step": 1396224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7131553896419136e-05,
+      "loss": 3.9299,
+      "step": 1396736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.712318432771235e-05,
+      "loss": 3.9185,
+      "step": 1397248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.711479838020183e-05,
+      "loss": 3.9339,
+      "step": 1397760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.710641243269131e-05,
+      "loss": 3.9135,
+      "step": 1398272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7098042863984518e-05,
+      "loss": 3.9183,
+      "step": 1398784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7089656916473998e-05,
+      "loss": 3.9302,
+      "step": 1399296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7081270968963478e-05,
+      "loss": 3.9269,
+      "step": 1399808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7072885021452958e-05,
+      "loss": 3.9218,
+      "step": 1400320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.706449907394244e-05,
+      "loss": 3.918,
+      "step": 1400832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.705611312643192e-05,
+      "loss": 3.9257,
+      "step": 1401344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.70477271789214e-05,
+      "loss": 3.9095,
+      "step": 1401856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.703934123141088e-05,
+      "loss": 3.9319,
+      "step": 1402368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7030955283900354e-05,
+      "loss": 3.9246,
+      "step": 1402880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.702258571519357e-05,
+      "loss": 3.933,
+      "step": 1403392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7014216146486783e-05,
+      "loss": 3.9285,
+      "step": 1403904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7005830198976263e-05,
+      "loss": 3.9162,
+      "step": 1404416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6997444251465743e-05,
+      "loss": 3.921,
+      "step": 1404928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6989058303955222e-05,
+      "loss": 3.9287,
+      "step": 1405440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6980672356444702e-05,
+      "loss": 3.9139,
+      "step": 1405952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.697228640893418e-05,
+      "loss": 3.9182,
+      "step": 1406464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.696390046142366e-05,
+      "loss": 3.9298,
+      "step": 1406976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.695551451391314e-05,
+      "loss": 3.924,
+      "step": 1407488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.694712856640262e-05,
+      "loss": 3.911,
+      "step": 1408000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.69387426188921e-05,
+      "loss": 3.9118,
+      "step": 1408512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.693035667138158e-05,
+      "loss": 3.9122,
+      "step": 1409024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6921987102674788e-05,
+      "loss": 3.9324,
+      "step": 1409536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.691360115516427e-05,
+      "loss": 3.9291,
+      "step": 1410048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.690523158645748e-05,
+      "loss": 3.9187,
+      "step": 1410560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.689684563894696e-05,
+      "loss": 3.9268,
+      "step": 1411072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.688845969143644e-05,
+      "loss": 3.923,
+      "step": 1411584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.688007374392592e-05,
+      "loss": 3.9309,
+      "step": 1412096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.68716877964154e-05,
+      "loss": 3.9187,
+      "step": 1412608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.686330184890488e-05,
+      "loss": 3.9281,
+      "step": 1413120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6854915901394363e-05,
+      "loss": 3.926,
+      "step": 1413632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6846529953883843e-05,
+      "loss": 3.9273,
+      "step": 1414144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6838144006373323e-05,
+      "loss": 3.9206,
+      "step": 1414656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6829774437666532e-05,
+      "loss": 3.9259,
+      "step": 1415168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6821388490156012e-05,
+      "loss": 3.9278,
+      "step": 1415680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6813002542645492e-05,
+      "loss": 3.9268,
+      "step": 1416192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6804616595134972e-05,
+      "loss": 3.9133,
+      "step": 1416704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6796230647624456e-05,
+      "loss": 3.921,
+      "step": 1417216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6787844700113936e-05,
+      "loss": 3.9196,
+      "step": 1417728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6779458752603416e-05,
+      "loss": 3.9263,
+      "step": 1418240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.677107280509289e-05,
+      "loss": 3.9278,
+      "step": 1418752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6762703236386105e-05,
+      "loss": 3.909,
+      "step": 1419264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6754317288875585e-05,
+      "loss": 3.9233,
+      "step": 1419776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6745931341365065e-05,
+      "loss": 3.9195,
+      "step": 1420288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.673754539385454e-05,
+      "loss": 3.9263,
+      "step": 1420800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6729175825147757e-05,
+      "loss": 3.9115,
+      "step": 1421312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6720789877637237e-05,
+      "loss": 3.9217,
+      "step": 1421824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.671240393012671e-05,
+      "loss": 3.9322,
+      "step": 1422336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6704017982616193e-05,
+      "loss": 3.9207,
+      "step": 1422848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6695632035105673e-05,
+      "loss": 3.9034,
+      "step": 1423360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6687246087595153e-05,
+      "loss": 3.9212,
+      "step": 1423872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6678876518888362e-05,
+      "loss": 3.9261,
+      "step": 1424384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6670490571377842e-05,
+      "loss": 3.9268,
+      "step": 1424896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6662104623867322e-05,
+      "loss": 3.9232,
+      "step": 1425408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6653735055160538e-05,
+      "loss": 3.9141,
+      "step": 1425920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6645349107650015e-05,
+      "loss": 3.9152,
+      "step": 1426432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6636963160139495e-05,
+      "loss": 3.9266,
+      "step": 1426944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6628577212628975e-05,
+      "loss": 3.9084,
+      "step": 1427456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6620191265118455e-05,
+      "loss": 3.9147,
+      "step": 1427968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6611821696411664e-05,
+      "loss": 3.9214,
+      "step": 1428480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6603435748901147e-05,
+      "loss": 3.9227,
+      "step": 1428992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6595049801390627e-05,
+      "loss": 3.9008,
+      "step": 1429504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6586663853880107e-05,
+      "loss": 3.9203,
+      "step": 1430016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6578277906369587e-05,
+      "loss": 3.9083,
+      "step": 1430528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6569891958859067e-05,
+      "loss": 3.9206,
+      "step": 1431040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6561506011348547e-05,
+      "loss": 3.9166,
+      "step": 1431552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6553120063838027e-05,
+      "loss": 3.9099,
+      "step": 1432064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6544750495131236e-05,
+      "loss": 3.9255,
+      "step": 1432576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.653638092642445e-05,
+      "loss": 3.9253,
+      "step": 1433088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.652799497891393e-05,
+      "loss": 3.9121,
+      "step": 1433600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.651960903140341e-05,
+      "loss": 3.9201,
+      "step": 1434112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6511223083892888e-05,
+      "loss": 3.9199,
+      "step": 1434624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6502837136382368e-05,
+      "loss": 3.9202,
+      "step": 1435136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6494451188871848e-05,
+      "loss": 3.9073,
+      "step": 1435648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.648608162016506e-05,
+      "loss": 3.9251,
+      "step": 1436160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.647769567265454e-05,
+      "loss": 3.9173,
+      "step": 1436672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.646930972514402e-05,
+      "loss": 3.922,
+      "step": 1437184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.64609237776335e-05,
+      "loss": 3.9175,
+      "step": 1437696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.645253783012298e-05,
+      "loss": 3.9217,
+      "step": 1438208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.644416826141619e-05,
+      "loss": 3.9097,
+      "step": 1438720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6435782313905673e-05,
+      "loss": 3.9144,
+      "step": 1439232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6427396366395153e-05,
+      "loss": 3.9034,
+      "step": 1439744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6419010418884633e-05,
+      "loss": 3.9327,
+      "step": 1440256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6410624471374113e-05,
+      "loss": 3.9235,
+      "step": 1440768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6402254902667322e-05,
+      "loss": 3.9232,
+      "step": 1441280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6393868955156802e-05,
+      "loss": 3.9104,
+      "step": 1441792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6385483007646282e-05,
+      "loss": 3.9274,
+      "step": 1442304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6377097060135765e-05,
+      "loss": 3.9254,
+      "step": 1442816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6368711112625245e-05,
+      "loss": 3.9136,
+      "step": 1443328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6360325165114725e-05,
+      "loss": 3.9166,
+      "step": 1443840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6351939217604198e-05,
+      "loss": 3.9238,
+      "step": 1444352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6343553270093678e-05,
+      "loss": 3.92,
+      "step": 1444864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6335183701386894e-05,
+      "loss": 3.927,
+      "step": 1445376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6326797753876374e-05,
+      "loss": 3.9096,
+      "step": 1445888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6318428185169587e-05,
+      "loss": 3.9235,
+      "step": 1446400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6310042237659066e-05,
+      "loss": 3.9199,
+      "step": 1446912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6301656290148546e-05,
+      "loss": 3.9124,
+      "step": 1447424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6293286721441756e-05,
+      "loss": 3.9327,
+      "step": 1447936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6284900773931235e-05,
+      "loss": 3.9127,
+      "step": 1448448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.627651482642072e-05,
+      "loss": 3.9186,
+      "step": 1448960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6268145257713928e-05,
+      "loss": 3.922,
+      "step": 1449472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6259759310203408e-05,
+      "loss": 3.9175,
+      "step": 1449984
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.054660797119141,
+      "eval_runtime": 292.6604,
+      "eval_samples_per_second": 1303.869,
+      "eval_steps_per_second": 40.747,
+      "step": 1450080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6251373362692888e-05,
+      "loss": 3.9176,
+      "step": 1450496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6242987415182368e-05,
+      "loss": 3.9186,
+      "step": 1451008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6234601467671848e-05,
+      "loss": 3.9257,
+      "step": 1451520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6226215520161324e-05,
+      "loss": 3.9227,
+      "step": 1452032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6217829572650804e-05,
+      "loss": 3.9283,
+      "step": 1452544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6209443625140284e-05,
+      "loss": 3.912,
+      "step": 1453056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6201057677629764e-05,
+      "loss": 3.9162,
+      "step": 1453568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6192671730119244e-05,
+      "loss": 3.9059,
+      "step": 1454080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6184285782608724e-05,
+      "loss": 3.9284,
+      "step": 1454592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6175899835098204e-05,
+      "loss": 3.9135,
+      "step": 1455104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6167513887587687e-05,
+      "loss": 3.9251,
+      "step": 1455616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6159144318880896e-05,
+      "loss": 3.9244,
+      "step": 1456128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6150774750174106e-05,
+      "loss": 3.9165,
+      "step": 1456640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6142388802663586e-05,
+      "loss": 3.9067,
+      "step": 1457152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6134002855153066e-05,
+      "loss": 3.9077,
+      "step": 1457664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.612561690764255e-05,
+      "loss": 3.9082,
+      "step": 1458176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.611723096013203e-05,
+      "loss": 3.9179,
+      "step": 1458688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.610884501262151e-05,
+      "loss": 3.9091,
+      "step": 1459200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.610045906511099e-05,
+      "loss": 3.9129,
+      "step": 1459712
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.609207311760047e-05,
+      "loss": 3.9342,
+      "step": 1460224
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.608368717008995e-05,
+      "loss": 3.9164,
+      "step": 1460736
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.607530122257943e-05,
+      "loss": 3.9156,
+      "step": 1461248
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.606691527506891e-05,
+      "loss": 3.9269,
+      "step": 1461760
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.605854570636212e-05,
+      "loss": 3.92,
+      "step": 1462272
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.60501597588516e-05,
+      "loss": 3.9121,
+      "step": 1462784
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.604177381134108e-05,
+      "loss": 3.9052,
+      "step": 1463296
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.603338786383056e-05,
+      "loss": 3.9223,
+      "step": 1463808
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6025001916320034e-05,
+      "loss": 3.9053,
+      "step": 1464320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6016615968809514e-05,
+      "loss": 3.9084,
+      "step": 1464832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6008230021298997e-05,
+      "loss": 3.9132,
+      "step": 1465344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5999844073788477e-05,
+      "loss": 3.9088,
+      "step": 1465856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5991474505081686e-05,
+      "loss": 3.9242,
+      "step": 1466368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5983088557571166e-05,
+      "loss": 3.9178,
+      "step": 1466880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5974718988864382e-05,
+      "loss": 3.9074,
+      "step": 1467392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.596634942015759e-05,
+      "loss": 3.9208,
+      "step": 1467904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5957963472647075e-05,
+      "loss": 3.9261,
+      "step": 1468416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5949577525136555e-05,
+      "loss": 3.9057,
+      "step": 1468928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5941191577626035e-05,
+      "loss": 3.9182,
+      "step": 1469440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5932805630115508e-05,
+      "loss": 3.9042,
+      "step": 1469952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5924419682604988e-05,
+      "loss": 3.9187,
+      "step": 1470464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5916033735094468e-05,
+      "loss": 3.9078,
+      "step": 1470976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.590764778758395e-05,
+      "loss": 3.9295,
+      "step": 1471488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.589926184007343e-05,
+      "loss": 3.9183,
+      "step": 1472000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.589089227136664e-05,
+      "loss": 3.9154,
+      "step": 1472512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.588250632385612e-05,
+      "loss": 3.9213,
+      "step": 1473024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.58741203763456e-05,
+      "loss": 3.9103,
+      "step": 1473536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.586573442883508e-05,
+      "loss": 3.9224,
+      "step": 1474048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.585734848132456e-05,
+      "loss": 3.9082,
+      "step": 1474560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5848962533814043e-05,
+      "loss": 3.9064,
+      "step": 1475072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5840592965107252e-05,
+      "loss": 3.9233,
+      "step": 1475584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5832207017596732e-05,
+      "loss": 3.9171,
+      "step": 1476096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5823821070086212e-05,
+      "loss": 3.9174,
+      "step": 1476608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5815435122575692e-05,
+      "loss": 3.9061,
+      "step": 1477120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5807065553868905e-05,
+      "loss": 3.9149,
+      "step": 1477632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5798679606358385e-05,
+      "loss": 3.9019,
+      "step": 1478144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5790293658847865e-05,
+      "loss": 3.9217,
+      "step": 1478656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5781907711337345e-05,
+      "loss": 3.9156,
+      "step": 1479168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5773538142630554e-05,
+      "loss": 3.9269,
+      "step": 1479680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5765152195120034e-05,
+      "loss": 3.9189,
+      "step": 1480192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5756766247609514e-05,
+      "loss": 3.9106,
+      "step": 1480704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5748380300098997e-05,
+      "loss": 3.9066,
+      "step": 1481216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5739994352588477e-05,
+      "loss": 3.92,
+      "step": 1481728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5731608405077957e-05,
+      "loss": 3.9067,
+      "step": 1482240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5723222457567437e-05,
+      "loss": 3.9093,
+      "step": 1482752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5714836510056917e-05,
+      "loss": 3.918,
+      "step": 1483264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5706450562546397e-05,
+      "loss": 3.9173,
+      "step": 1483776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5698080993839606e-05,
+      "loss": 3.8994,
+      "step": 1484288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.568969504632909e-05,
+      "loss": 3.9036,
+      "step": 1484800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.568130909881857e-05,
+      "loss": 3.9009,
+      "step": 1485312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5672923151308042e-05,
+      "loss": 3.926,
+      "step": 1485824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5664553582601258e-05,
+      "loss": 3.9187,
+      "step": 1486336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5656167635090738e-05,
+      "loss": 3.9109,
+      "step": 1486848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5647781687580218e-05,
+      "loss": 3.9171,
+      "step": 1487360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5639395740069695e-05,
+      "loss": 3.9107,
+      "step": 1487872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.563102617136291e-05,
+      "loss": 3.9274,
+      "step": 1488384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.562265660265612e-05,
+      "loss": 3.9036,
+      "step": 1488896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.56142706551456e-05,
+      "loss": 3.9263,
+      "step": 1489408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.560588470763508e-05,
+      "loss": 3.9122,
+      "step": 1489920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.559749876012456e-05,
+      "loss": 3.918,
+      "step": 1490432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5589112812614043e-05,
+      "loss": 3.9135,
+      "step": 1490944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5580726865103516e-05,
+      "loss": 3.9127,
+      "step": 1491456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5572340917592996e-05,
+      "loss": 3.9226,
+      "step": 1491968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5563954970082476e-05,
+      "loss": 3.9156,
+      "step": 1492480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5555569022571956e-05,
+      "loss": 3.9053,
+      "step": 1492992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5547183075061436e-05,
+      "loss": 3.9152,
+      "step": 1493504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.553879712755092e-05,
+      "loss": 3.9054,
+      "step": 1494016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.55304111800404e-05,
+      "loss": 3.9219,
+      "step": 1494528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5522041611333608e-05,
+      "loss": 3.9176,
+      "step": 1495040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5513655663823088e-05,
+      "loss": 3.9014,
+      "step": 1495552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5505269716312568e-05,
+      "loss": 3.9171,
+      "step": 1496064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5496883768802048e-05,
+      "loss": 3.9073,
+      "step": 1496576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.548851420009526e-05,
+      "loss": 3.914,
+      "step": 1497088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.548012825258474e-05,
+      "loss": 3.9041,
+      "step": 1497600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.547174230507422e-05,
+      "loss": 3.9137,
+      "step": 1498112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.54633563575637e-05,
+      "loss": 3.9232,
+      "step": 1498624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.545498678885691e-05,
+      "loss": 3.9089,
+      "step": 1499136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.544660084134639e-05,
+      "loss": 3.9001,
+      "step": 1499648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.543821489383587e-05,
+      "loss": 3.911,
+      "step": 1500160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5429828946325353e-05,
+      "loss": 3.9082,
+      "step": 1500672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5421442998814833e-05,
+      "loss": 3.9254,
+      "step": 1501184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5413073430108042e-05,
+      "loss": 3.9121,
+      "step": 1501696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5404687482597522e-05,
+      "loss": 3.907,
+      "step": 1502208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5396301535087002e-05,
+      "loss": 3.9062,
+      "step": 1502720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.538791558757648e-05,
+      "loss": 3.9202,
+      "step": 1503232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5379546018869694e-05,
+      "loss": 3.8963,
+      "step": 1503744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5371160071359174e-05,
+      "loss": 3.9076,
+      "step": 1504256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5362790502652383e-05,
+      "loss": 3.9106,
+      "step": 1504768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5354404555141863e-05,
+      "loss": 3.9173,
+      "step": 1505280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5346018607631343e-05,
+      "loss": 3.8905,
+      "step": 1505792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5337632660120823e-05,
+      "loss": 3.9107,
+      "step": 1506304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5329246712610306e-05,
+      "loss": 3.906,
+      "step": 1506816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5320860765099786e-05,
+      "loss": 3.9046,
+      "step": 1507328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5312474817589266e-05,
+      "loss": 3.912,
+      "step": 1507840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5304105248882475e-05,
+      "loss": 3.8995,
+      "step": 1508352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5295735680175685e-05,
+      "loss": 3.9208,
+      "step": 1508864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5287349732665168e-05,
+      "loss": 3.9141,
+      "step": 1509376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5278980163958377e-05,
+      "loss": 3.9048,
+      "step": 1509888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5270594216447857e-05,
+      "loss": 3.9123,
+      "step": 1510400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5262208268937337e-05,
+      "loss": 3.91,
+      "step": 1510912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5253822321426817e-05,
+      "loss": 3.9122,
+      "step": 1511424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5245436373916297e-05,
+      "loss": 3.8928,
+      "step": 1511936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5237050426405777e-05,
+      "loss": 3.9207,
+      "step": 1512448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.522866447889526e-05,
+      "loss": 3.9051,
+      "step": 1512960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.522027853138474e-05,
+      "loss": 3.9112,
+      "step": 1513472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.521189258387422e-05,
+      "loss": 3.9089,
+      "step": 1513984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.520352301516743e-05,
+      "loss": 3.9161,
+      "step": 1514496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.519513706765691e-05,
+      "loss": 3.9051,
+      "step": 1515008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.518675112014639e-05,
+      "loss": 3.903,
+      "step": 1515520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.517836517263587e-05,
+      "loss": 3.8963,
+      "step": 1516032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5169979225125352e-05,
+      "loss": 3.9212,
+      "step": 1516544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5161593277614825e-05,
+      "loss": 3.9142,
+      "step": 1517056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5153207330104305e-05,
+      "loss": 3.9155,
+      "step": 1517568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5144821382593785e-05,
+      "loss": 3.9043,
+      "step": 1518080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5136435435083265e-05,
+      "loss": 3.9187,
+      "step": 1518592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5128065866376478e-05,
+      "loss": 3.9175,
+      "step": 1519104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5119679918865958e-05,
+      "loss": 3.9052,
+      "step": 1519616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5111293971355438e-05,
+      "loss": 3.9073,
+      "step": 1520128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5102908023844918e-05,
+      "loss": 3.9145,
+      "step": 1520640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5094522076334398e-05,
+      "loss": 3.916,
+      "step": 1521152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5086152507627607e-05,
+      "loss": 3.9199,
+      "step": 1521664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.507776656011709e-05,
+      "loss": 3.9006,
+      "step": 1522176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.50693969914103e-05,
+      "loss": 3.9164,
+      "step": 1522688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.506101104389978e-05,
+      "loss": 3.9083,
+      "step": 1523200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.505262509638926e-05,
+      "loss": 3.9075,
+      "step": 1523712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.504423914887874e-05,
+      "loss": 3.9184,
+      "step": 1524224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.503585320136822e-05,
+      "loss": 3.9081,
+      "step": 1524736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.50274672538577e-05,
+      "loss": 3.9132,
+      "step": 1525248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5019081306347182e-05,
+      "loss": 3.9081,
+      "step": 1525760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5010695358836662e-05,
+      "loss": 3.9141,
+      "step": 1526272
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.050495624542236,
+      "eval_runtime": 291.1309,
+      "eval_samples_per_second": 1310.72,
+      "eval_steps_per_second": 40.961,
+      "step": 1526400
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.50023421689336e-05,
+      "loss": 3.9093,
+      "step": 1526784
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4993956221423084e-05,
+      "loss": 3.9136,
+      "step": 1527296
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4985570273912564e-05,
+      "loss": 3.9158,
+      "step": 1527808
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4977184326402044e-05,
+      "loss": 3.9117,
+      "step": 1528320
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4968798378891524e-05,
+      "loss": 3.9222,
+      "step": 1528832
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4960412431381004e-05,
+      "loss": 3.9032,
+      "step": 1529344
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4952026483870484e-05,
+      "loss": 3.9104,
+      "step": 1529856
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4943640536359964e-05,
+      "loss": 3.9009,
+      "step": 1530368
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4935270967653173e-05,
+      "loss": 3.9168,
+      "step": 1530880
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4926885020142653e-05,
+      "loss": 3.9107,
+      "step": 1531392
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4918499072632136e-05,
+      "loss": 3.9142,
+      "step": 1531904
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4910113125121616e-05,
+      "loss": 3.9133,
+      "step": 1532416
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4901743556414825e-05,
+      "loss": 3.9083,
+      "step": 1532928
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4893357608904305e-05,
+      "loss": 3.9009,
+      "step": 1533440
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4884971661393785e-05,
+      "loss": 3.9047,
+      "step": 1533952
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4876585713883265e-05,
+      "loss": 3.8964,
+      "step": 1534464
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4868199766372745e-05,
+      "loss": 3.911,
+      "step": 1534976
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4859813818862225e-05,
+      "loss": 3.9012,
+      "step": 1535488
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4851427871351705e-05,
+      "loss": 3.9014,
+      "step": 1536000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4843041923841185e-05,
+      "loss": 3.9317,
+      "step": 1536512
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4834672355134394e-05,
+      "loss": 3.9053,
+      "step": 1537024
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4826286407623874e-05,
+      "loss": 3.9121,
+      "step": 1537536
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4817900460113354e-05,
+      "loss": 3.9168,
+      "step": 1538048
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4809514512602837e-05,
+      "loss": 3.9118,
+      "step": 1538560
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4801128565092317e-05,
+      "loss": 3.9058,
+      "step": 1539072
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4792758996385526e-05,
+      "loss": 3.897,
+      "step": 1539584
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4784373048875006e-05,
+      "loss": 3.9168,
+      "step": 1540096
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.4776003480168215e-05,
+      "loss": 3.8928,
+      "step": 1540608
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.47676175326577e-05,
+      "loss": 3.9037,
+      "step": 1541120
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.475923158514718e-05,
+      "loss": 3.9007,
+      "step": 1541632
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.475084563763666e-05,
+      "loss": 3.9047,
+      "step": 1542144
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.474245969012614e-05,
+      "loss": 3.9148,
+      "step": 1542656
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.473407374261562e-05,
+      "loss": 3.9124,
+      "step": 1543168
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4725704173908827e-05,
+      "loss": 3.9061,
+      "step": 1543680
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.471733460520204e-05,
+      "loss": 3.9133,
+      "step": 1544192
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.470894865769152e-05,
+      "loss": 3.9134,
+      "step": 1544704
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4700562710181e-05,
+      "loss": 3.901,
+      "step": 1545216
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.469217676267048e-05,
+      "loss": 3.9109,
+      "step": 1545728
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.468379081515996e-05,
+      "loss": 3.8966,
+      "step": 1546240
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.467540486764944e-05,
+      "loss": 3.912,
+      "step": 1546752
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.466701892013892e-05,
+      "loss": 3.8963,
+      "step": 1547264
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.46586329726284e-05,
+      "loss": 3.9245,
+      "step": 1547776
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.465024702511788e-05,
+      "loss": 3.9103,
+      "step": 1548288
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.464186107760736e-05,
+      "loss": 3.9074,
+      "step": 1548800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.463347513009684e-05,
+      "loss": 3.9126,
+      "step": 1549312
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.462508918258632e-05,
+      "loss": 3.9042,
+      "step": 1549824
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.461671961387953e-05,
+      "loss": 3.915,
+      "step": 1550336
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.460833366636901e-05,
+      "loss": 3.9005,
+      "step": 1550848
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4599947718858492e-05,
+      "loss": 3.8967,
+      "step": 1551360
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4591561771347972e-05,
+      "loss": 3.9174,
+      "step": 1551872
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.458319220264118e-05,
+      "loss": 3.9059,
+      "step": 1552384
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.457480625513066e-05,
+      "loss": 3.9099,
+      "step": 1552896
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.456642030762014e-05,
+      "loss": 3.899,
+      "step": 1553408
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4558050738913353e-05,
+      "loss": 3.9079,
+      "step": 1553920
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4549664791402833e-05,
+      "loss": 3.8976,
+      "step": 1554432
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4541278843892313e-05,
+      "loss": 3.9125,
+      "step": 1554944
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4532892896381793e-05,
+      "loss": 3.9071,
+      "step": 1555456
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4524506948871273e-05,
+      "loss": 3.9211,
+      "step": 1555968
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4516121001360753e-05,
+      "loss": 3.9097,
+      "step": 1556480
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.450773505385023e-05,
+      "loss": 3.9011,
+      "step": 1556992
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4499365485143446e-05,
+      "loss": 3.9024,
+      "step": 1557504
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4490979537632925e-05,
+      "loss": 3.9111,
+      "step": 1558016
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4482593590122405e-05,
+      "loss": 3.8952,
+      "step": 1558528
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4474207642611882e-05,
+      "loss": 3.9036,
+      "step": 1559040
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4465838073905094e-05,
+      "loss": 3.9085,
+      "step": 1559552
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4457468505198307e-05,
+      "loss": 3.9135,
+      "step": 1560064
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4449082557687787e-05,
+      "loss": 3.8918,
+      "step": 1560576
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4440696610177267e-05,
+      "loss": 3.8981,
+      "step": 1561088
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4432327041470476e-05,
+      "loss": 3.8935,
+      "step": 1561600
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4423941093959956e-05,
+      "loss": 3.9161,
+      "step": 1562112
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4415555146449436e-05,
+      "loss": 3.9094,
+      "step": 1562624
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4407169198938916e-05,
+      "loss": 3.9055,
+      "step": 1563136
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.43987832514284e-05,
+      "loss": 3.9083,
+      "step": 1563648
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.439039730391788e-05,
+      "loss": 3.9026,
+      "step": 1564160
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4382011356407356e-05,
+      "loss": 3.9233,
+      "step": 1564672
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4373625408896836e-05,
+      "loss": 3.8958,
+      "step": 1565184
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4365239461386316e-05,
+      "loss": 3.9134,
+      "step": 1565696
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4356853513875796e-05,
+      "loss": 3.9101,
+      "step": 1566208
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4348467566365276e-05,
+      "loss": 3.9088,
+      "step": 1566720
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4340081618854755e-05,
+      "loss": 3.9099,
+      "step": 1567232
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4331695671344235e-05,
+      "loss": 3.908,
+      "step": 1567744
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4323309723833715e-05,
+      "loss": 3.9109,
+      "step": 1568256
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4314923776323195e-05,
+      "loss": 3.9092,
+      "step": 1568768
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4306537828812675e-05,
+      "loss": 3.8949,
+      "step": 1569280
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4298168260105884e-05,
+      "loss": 3.9104,
+      "step": 1569792
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4289782312595368e-05,
+      "loss": 3.8988,
+      "step": 1570304
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4281396365084848e-05,
+      "loss": 3.9132,
+      "step": 1570816
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.4273010417574328e-05,
+      "loss": 3.9087,
+      "step": 1571328
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.426465722767127e-05,
+      "loss": 3.8948,
+      "step": 1571840
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.425627128016075e-05,
+      "loss": 3.909,
+      "step": 1572352
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.424788533265023e-05,
+      "loss": 3.9017,
+      "step": 1572864
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.423949938513971e-05,
+      "loss": 3.9097,
+      "step": 1573376
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.423111343762919e-05,
+      "loss": 3.8918,
+      "step": 1573888
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.422272749011867e-05,
+      "loss": 3.9075,
+      "step": 1574400
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.421434154260815e-05,
+      "loss": 3.919,
+      "step": 1574912
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.420595559509763e-05,
+      "loss": 3.9003,
+      "step": 1575424
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4197586026390838e-05,
+      "loss": 3.888,
+      "step": 1575936
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4189216457684054e-05,
+      "loss": 3.9047,
+      "step": 1576448
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.418083051017353e-05,
+      "loss": 3.9042,
+      "step": 1576960
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.417244456266301e-05,
+      "loss": 3.9134,
+      "step": 1577472
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.416405861515249e-05,
+      "loss": 3.9079,
+      "step": 1577984
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.415567266764197e-05,
+      "loss": 3.9003,
+      "step": 1578496
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.414730309893518e-05,
+      "loss": 3.8983,
+      "step": 1579008
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4138917151424663e-05,
+      "loss": 3.9122,
+      "step": 1579520
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4130531203914143e-05,
+      "loss": 3.8887,
+      "step": 1580032
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4122145256403623e-05,
+      "loss": 3.8988,
+      "step": 1580544
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4113775687696832e-05,
+      "loss": 3.9037,
+      "step": 1581056
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4105389740186312e-05,
+      "loss": 3.9092,
+      "step": 1581568
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4097003792675792e-05,
+      "loss": 3.8853,
+      "step": 1582080
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.408861784516527e-05,
+      "loss": 3.9047,
+      "step": 1582592
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4080231897654755e-05,
+      "loss": 3.8961,
+      "step": 1583104
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4071845950144235e-05,
+      "loss": 3.9007,
+      "step": 1583616
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.406346000263371e-05,
+      "loss": 3.9017,
+      "step": 1584128
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.405507405512319e-05,
+      "loss": 3.8961,
+      "step": 1584640
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.404668810761267e-05,
+      "loss": 3.9089,
+      "step": 1585152
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.403830216010215e-05,
+      "loss": 3.9098,
+      "step": 1585664
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4029948970199096e-05,
+      "loss": 3.8967,
+      "step": 1586176
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4021563022688576e-05,
+      "loss": 3.9066,
+      "step": 1586688
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4013177075178056e-05,
+      "loss": 3.8992,
+      "step": 1587200
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.4004791127667536e-05,
+      "loss": 3.9091,
+      "step": 1587712
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3996405180157013e-05,
+      "loss": 3.89,
+      "step": 1588224
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3988019232646493e-05,
+      "loss": 3.9146,
+      "step": 1588736
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3979633285135976e-05,
+      "loss": 3.8933,
+      "step": 1589248
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3971247337625456e-05,
+      "loss": 3.9056,
+      "step": 1589760
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3962861390114936e-05,
+      "loss": 3.8949,
+      "step": 1590272
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3954475442604416e-05,
+      "loss": 3.9093,
+      "step": 1590784
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3946089495093893e-05,
+      "loss": 3.9008,
+      "step": 1591296
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3937736305190838e-05,
+      "loss": 3.8973,
+      "step": 1591808
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3929350357680318e-05,
+      "loss": 3.8866,
+      "step": 1592320
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3920964410169798e-05,
+      "loss": 3.9179,
+      "step": 1592832
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3912578462659277e-05,
+      "loss": 3.9023,
+      "step": 1593344
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3904192515148757e-05,
+      "loss": 3.91,
+      "step": 1593856
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3895806567638237e-05,
+      "loss": 3.8966,
+      "step": 1594368
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3887420620127717e-05,
+      "loss": 3.9107,
+      "step": 1594880
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3879034672617194e-05,
+      "loss": 3.9128,
+      "step": 1595392
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3870648725106677e-05,
+      "loss": 3.8964,
+      "step": 1595904
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.386227915639989e-05,
+      "loss": 3.9005,
+      "step": 1596416
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3853893208889366e-05,
+      "loss": 3.9061,
+      "step": 1596928
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3845507261378846e-05,
+      "loss": 3.9089,
+      "step": 1597440
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3837121313868326e-05,
+      "loss": 3.9132,
+      "step": 1597952
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3828735366357806e-05,
+      "loss": 3.8937,
+      "step": 1598464
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3820349418847286e-05,
+      "loss": 3.9069,
+      "step": 1598976
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.381196347133677e-05,
+      "loss": 3.8999,
+      "step": 1599488
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.380357752382625e-05,
+      "loss": 3.9032,
+      "step": 1600000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3795191576315726e-05,
+      "loss": 3.9093,
+      "step": 1600512
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.378682200760894e-05,
+      "loss": 3.8967,
+      "step": 1601024
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3778452438902148e-05,
+      "loss": 3.9112,
+      "step": 1601536
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.377006649139163e-05,
+      "loss": 3.8985,
+      "step": 1602048
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.376169692268484e-05,
+      "loss": 3.9112,
+      "step": 1602560
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.047706604003906,
+      "eval_runtime": 283.46,
+      "eval_samples_per_second": 1346.19,
+      "eval_steps_per_second": 42.069,
+      "step": 1602720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.375331097517432e-05,
+      "loss": 3.8926,
+      "step": 1603072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.37449250276638e-05,
+      "loss": 3.9077,
+      "step": 1603584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.373653908015328e-05,
+      "loss": 3.9065,
+      "step": 1604096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.372815313264276e-05,
+      "loss": 3.9037,
+      "step": 1604608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.371976718513224e-05,
+      "loss": 3.9123,
+      "step": 1605120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3711381237621723e-05,
+      "loss": 3.8988,
+      "step": 1605632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.37029952901112e-05,
+      "loss": 3.8998,
+      "step": 1606144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.369460934260068e-05,
+      "loss": 3.8922,
+      "step": 1606656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.368622339509016e-05,
+      "loss": 3.9081,
+      "step": 1607168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.367783744757964e-05,
+      "loss": 3.9065,
+      "step": 1607680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.366945150006912e-05,
+      "loss": 3.904,
+      "step": 1608192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.36610655525586e-05,
+      "loss": 3.9072,
+      "step": 1608704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3652695983851812e-05,
+      "loss": 3.8978,
+      "step": 1609216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3644310036341292e-05,
+      "loss": 3.8952,
+      "step": 1609728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3635924088830772e-05,
+      "loss": 3.8981,
+      "step": 1610240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3627538141320252e-05,
+      "loss": 3.8887,
+      "step": 1610752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.361915219380973e-05,
+      "loss": 3.8996,
+      "step": 1611264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.361076624629921e-05,
+      "loss": 3.8924,
+      "step": 1611776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.360238029878869e-05,
+      "loss": 3.8954,
+      "step": 1612288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.359399435127817e-05,
+      "loss": 3.9302,
+      "step": 1612800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.358560840376765e-05,
+      "loss": 3.8951,
+      "step": 1613312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.357723883506086e-05,
+      "loss": 3.9052,
+      "step": 1613824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.356885288755034e-05,
+      "loss": 3.9068,
+      "step": 1614336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.356046694003982e-05,
+      "loss": 3.9026,
+      "step": 1614848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.35520809925293e-05,
+      "loss": 3.8945,
+      "step": 1615360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.354369504501878e-05,
+      "loss": 3.895,
+      "step": 1615872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.353530909750826e-05,
+      "loss": 3.9057,
+      "step": 1616384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3526939528801473e-05,
+      "loss": 3.8863,
+      "step": 1616896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3518553581290953e-05,
+      "loss": 3.8945,
+      "step": 1617408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3510167633780433e-05,
+      "loss": 3.8924,
+      "step": 1617920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.350178168626991e-05,
+      "loss": 3.9002,
+      "step": 1618432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3493395738759393e-05,
+      "loss": 3.9031,
+      "step": 1618944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3485009791248873e-05,
+      "loss": 3.9057,
+      "step": 1619456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3476623843738353e-05,
+      "loss": 3.8983,
+      "step": 1619968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3468254275031562e-05,
+      "loss": 3.9044,
+      "step": 1620480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3459868327521042e-05,
+      "loss": 3.9035,
+      "step": 1620992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3451482380010522e-05,
+      "loss": 3.893,
+      "step": 1621504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.34430964325e-05,
+      "loss": 3.9077,
+      "step": 1622016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3434726863793214e-05,
+      "loss": 3.8864,
+      "step": 1622528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3426340916282694e-05,
+      "loss": 3.9089,
+      "step": 1623040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3417954968772174e-05,
+      "loss": 3.8839,
+      "step": 1623552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3409569021261654e-05,
+      "loss": 3.9192,
+      "step": 1624064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3401183073751134e-05,
+      "loss": 3.903,
+      "step": 1624576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3392813505044343e-05,
+      "loss": 3.9003,
+      "step": 1625088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3384427557533826e-05,
+      "loss": 3.9044,
+      "step": 1625600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3376041610023306e-05,
+      "loss": 3.8988,
+      "step": 1626112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3367655662512786e-05,
+      "loss": 3.9032,
+      "step": 1626624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3359269715002266e-05,
+      "loss": 3.8915,
+      "step": 1627136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3350900146295475e-05,
+      "loss": 3.8839,
+      "step": 1627648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3342514198784955e-05,
+      "loss": 3.9163,
+      "step": 1628160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3334128251274435e-05,
+      "loss": 3.8987,
+      "step": 1628672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3325742303763915e-05,
+      "loss": 3.9066,
+      "step": 1629184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3317356356253395e-05,
+      "loss": 3.8924,
+      "step": 1629696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3308970408742875e-05,
+      "loss": 3.8969,
+      "step": 1630208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3300600840036088e-05,
+      "loss": 3.8906,
+      "step": 1630720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3292214892525564e-05,
+      "loss": 3.904,
+      "step": 1631232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3283828945015048e-05,
+      "loss": 3.8958,
+      "step": 1631744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3275442997504528e-05,
+      "loss": 3.9114,
+      "step": 1632256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3267057049994007e-05,
+      "loss": 3.9078,
+      "step": 1632768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3258671102483487e-05,
+      "loss": 3.8911,
+      "step": 1633280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3250285154972967e-05,
+      "loss": 3.8954,
+      "step": 1633792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3241899207462447e-05,
+      "loss": 3.9054,
+      "step": 1634304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3233529638755656e-05,
+      "loss": 3.8881,
+      "step": 1634816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.322514369124514e-05,
+      "loss": 3.8971,
+      "step": 1635328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.321675774373462e-05,
+      "loss": 3.9019,
+      "step": 1635840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3208371796224096e-05,
+      "loss": 3.9016,
+      "step": 1636352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3199985848713576e-05,
+      "loss": 3.8856,
+      "step": 1636864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.319161628000679e-05,
+      "loss": 3.8897,
+      "step": 1637376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.318323033249627e-05,
+      "loss": 3.8847,
+      "step": 1637888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.317484438498575e-05,
+      "loss": 3.9068,
+      "step": 1638400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.316645843747523e-05,
+      "loss": 3.8984,
+      "step": 1638912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.315810524757217e-05,
+      "loss": 3.8989,
+      "step": 1639424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.314971930006165e-05,
+      "loss": 3.9045,
+      "step": 1639936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.314133335255113e-05,
+      "loss": 3.893,
+      "step": 1640448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.313294740504061e-05,
+      "loss": 3.9164,
+      "step": 1640960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.312456145753009e-05,
+      "loss": 3.89,
+      "step": 1641472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.311617551001957e-05,
+      "loss": 3.9057,
+      "step": 1641984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.310778956250905e-05,
+      "loss": 3.9022,
+      "step": 1642496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.309940361499853e-05,
+      "loss": 3.8986,
+      "step": 1643008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3091034046291742e-05,
+      "loss": 3.9046,
+      "step": 1643520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.308264809878122e-05,
+      "loss": 3.8994,
+      "step": 1644032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3074278530074435e-05,
+      "loss": 3.8996,
+      "step": 1644544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3065892582563915e-05,
+      "loss": 3.904,
+      "step": 1645056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.305750663505339e-05,
+      "loss": 3.8848,
+      "step": 1645568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.304912068754287e-05,
+      "loss": 3.9046,
+      "step": 1646080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.304073474003235e-05,
+      "loss": 3.8909,
+      "step": 1646592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.303234879252183e-05,
+      "loss": 3.9055,
+      "step": 1647104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.302396284501131e-05,
+      "loss": 3.8971,
+      "step": 1647616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3015593276304524e-05,
+      "loss": 3.8876,
+      "step": 1648128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3007207328794004e-05,
+      "loss": 3.8993,
+      "step": 1648640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2998821381283484e-05,
+      "loss": 3.8981,
+      "step": 1649152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2990435433772964e-05,
+      "loss": 3.8992,
+      "step": 1649664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2982049486262444e-05,
+      "loss": 3.8849,
+      "step": 1650176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2973663538751923e-05,
+      "loss": 3.9041,
+      "step": 1650688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2965277591241403e-05,
+      "loss": 3.908,
+      "step": 1651200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2956908022534616e-05,
+      "loss": 3.8968,
+      "step": 1651712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2948522075024096e-05,
+      "loss": 3.8788,
+      "step": 1652224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2940136127513572e-05,
+      "loss": 3.8974,
+      "step": 1652736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2931750180003052e-05,
+      "loss": 3.8976,
+      "step": 1653248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2923364232492532e-05,
+      "loss": 3.9032,
+      "step": 1653760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2914978284982012e-05,
+      "loss": 3.9035,
+      "step": 1654272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2906608716275225e-05,
+      "loss": 3.8936,
+      "step": 1654784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2898239147568437e-05,
+      "loss": 3.8908,
+      "step": 1655296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2889853200057917e-05,
+      "loss": 3.9038,
+      "step": 1655808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2881467252547397e-05,
+      "loss": 3.8827,
+      "step": 1656320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2873081305036874e-05,
+      "loss": 3.8934,
+      "step": 1656832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2864695357526357e-05,
+      "loss": 3.8932,
+      "step": 1657344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2856309410015837e-05,
+      "loss": 3.9027,
+      "step": 1657856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2847923462505317e-05,
+      "loss": 3.8796,
+      "step": 1658368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2839537514994797e-05,
+      "loss": 3.8914,
+      "step": 1658880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2831167946288006e-05,
+      "loss": 3.8917,
+      "step": 1659392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2822781998777486e-05,
+      "loss": 3.8925,
+      "step": 1659904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2814396051266966e-05,
+      "loss": 3.895,
+      "step": 1660416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.280601010375645e-05,
+      "loss": 3.8861,
+      "step": 1660928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.279764053504966e-05,
+      "loss": 3.9033,
+      "step": 1661440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.278925458753914e-05,
+      "loss": 3.9011,
+      "step": 1661952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2780885018832347e-05,
+      "loss": 3.8875,
+      "step": 1662464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2772499071321827e-05,
+      "loss": 3.9007,
+      "step": 1662976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.276411312381131e-05,
+      "loss": 3.8896,
+      "step": 1663488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.275572717630079e-05,
+      "loss": 3.9032,
+      "step": 1664000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.274734122879027e-05,
+      "loss": 3.8869,
+      "step": 1664512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.273895528127975e-05,
+      "loss": 3.8993,
+      "step": 1665024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2730569333769227e-05,
+      "loss": 3.8862,
+      "step": 1665536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2722183386258707e-05,
+      "loss": 3.8975,
+      "step": 1666048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2713797438748187e-05,
+      "loss": 3.891,
+      "step": 1666560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2705427870041403e-05,
+      "loss": 3.9,
+      "step": 1667072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2697058301334612e-05,
+      "loss": 3.8915,
+      "step": 1667584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.268868873262782e-05,
+      "loss": 3.8964,
+      "step": 1668096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.26803027851173e-05,
+      "loss": 3.877,
+      "step": 1668608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.267191683760678e-05,
+      "loss": 3.912,
+      "step": 1669120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2663530890096264e-05,
+      "loss": 3.893,
+      "step": 1669632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2655144942585744e-05,
+      "loss": 3.9018,
+      "step": 1670144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2646758995075224e-05,
+      "loss": 3.886,
+      "step": 1670656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.26383730475647e-05,
+      "loss": 3.905,
+      "step": 1671168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.262998710005418e-05,
+      "loss": 3.9026,
+      "step": 1671680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.262160115254366e-05,
+      "loss": 3.8906,
+      "step": 1672192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.261321520503314e-05,
+      "loss": 3.8946,
+      "step": 1672704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.260482925752262e-05,
+      "loss": 3.8992,
+      "step": 1673216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2596443310012104e-05,
+      "loss": 3.9017,
+      "step": 1673728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2588057362501584e-05,
+      "loss": 3.9113,
+      "step": 1674240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.257967141499106e-05,
+      "loss": 3.8822,
+      "step": 1674752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.257128546748054e-05,
+      "loss": 3.9042,
+      "step": 1675264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.256289951997002e-05,
+      "loss": 3.8915,
+      "step": 1675776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2554529951263233e-05,
+      "loss": 3.9015,
+      "step": 1676288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2546160382556445e-05,
+      "loss": 3.8989,
+      "step": 1676800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2537774435045925e-05,
+      "loss": 3.8867,
+      "step": 1677312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2529388487535405e-05,
+      "loss": 3.9071,
+      "step": 1677824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2521002540024882e-05,
+      "loss": 3.8952,
+      "step": 1678336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2512616592514362e-05,
+      "loss": 3.9007,
+      "step": 1678848
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.043838977813721,
+      "eval_runtime": 295.9624,
+      "eval_samples_per_second": 1289.322,
+      "eval_steps_per_second": 40.292,
+      "step": 1679040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2504230645003842e-05,
+      "loss": 3.8917,
+      "step": 1679360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2495844697493322e-05,
+      "loss": 3.8953,
+      "step": 1679872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2487458749982805e-05,
+      "loss": 3.9027,
+      "step": 1680384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2479089181276014e-05,
+      "loss": 3.8995,
+      "step": 1680896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2470719612569227e-05,
+      "loss": 3.9026,
+      "step": 1681408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2462333665058703e-05,
+      "loss": 3.8959,
+      "step": 1681920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2453947717548183e-05,
+      "loss": 3.888,
+      "step": 1682432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2445561770037667e-05,
+      "loss": 3.8911,
+      "step": 1682944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2437175822527147e-05,
+      "loss": 3.8979,
+      "step": 1683456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2428806253820356e-05,
+      "loss": 3.9016,
+      "step": 1683968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2420420306309836e-05,
+      "loss": 3.8974,
+      "step": 1684480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2412034358799316e-05,
+      "loss": 3.902,
+      "step": 1684992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2403664790092528e-05,
+      "loss": 3.8868,
+      "step": 1685504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2395278842582008e-05,
+      "loss": 3.8919,
+      "step": 1686016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2386892895071488e-05,
+      "loss": 3.8927,
+      "step": 1686528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2378506947560968e-05,
+      "loss": 3.8821,
+      "step": 1687040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2370121000050448e-05,
+      "loss": 3.8951,
+      "step": 1687552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2361735052539928e-05,
+      "loss": 3.8837,
+      "step": 1688064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2353365483833137e-05,
+      "loss": 3.8882,
+      "step": 1688576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.234497953632262e-05,
+      "loss": 3.9235,
+      "step": 1689088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.23365935888121e-05,
+      "loss": 3.8922,
+      "step": 1689600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.232820764130158e-05,
+      "loss": 3.8949,
+      "step": 1690112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.231982169379106e-05,
+      "loss": 3.8998,
+      "step": 1690624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2311435746280537e-05,
+      "loss": 3.9033,
+      "step": 1691136
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2303049798770017e-05,
+      "loss": 3.8813,
+      "step": 1691648
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2294663851259497e-05,
+      "loss": 3.8911,
+      "step": 1692160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.228627790374898e-05,
+      "loss": 3.8941,
+      "step": 1692672
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.227789195623846e-05,
+      "loss": 3.8845,
+      "step": 1693184
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.226952238753167e-05,
+      "loss": 3.8896,
+      "step": 1693696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.226113644002115e-05,
+      "loss": 3.8852,
+      "step": 1694208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.225275049251063e-05,
+      "loss": 3.8942,
+      "step": 1694720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.224436454500011e-05,
+      "loss": 3.8994,
+      "step": 1695232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.223597859748959e-05,
+      "loss": 3.8966,
+      "step": 1695744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.222759264997907e-05,
+      "loss": 3.8913,
+      "step": 1696256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.221922308127228e-05,
+      "loss": 3.8971,
+      "step": 1696768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.221083713376176e-05,
+      "loss": 3.8987,
+      "step": 1697280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.220245118625124e-05,
+      "loss": 3.8861,
+      "step": 1697792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2194065238740718e-05,
+      "loss": 3.9039,
+      "step": 1698304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2185679291230198e-05,
+      "loss": 3.88,
+      "step": 1698816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.217729334371968e-05,
+      "loss": 3.899,
+      "step": 1699328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.216890739620916e-05,
+      "loss": 3.8784,
+      "step": 1699840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.216053782750237e-05,
+      "loss": 3.9096,
+      "step": 1700352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.215215187999185e-05,
+      "loss": 3.8934,
+      "step": 1700864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.214376593248133e-05,
+      "loss": 3.8966,
+      "step": 1701376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.213537998497081e-05,
+      "loss": 3.8969,
+      "step": 1701888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.212699403746029e-05,
+      "loss": 3.8935,
+      "step": 1702400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2118608089949773e-05,
+      "loss": 3.8989,
+      "step": 1702912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.211022214243925e-05,
+      "loss": 3.8869,
+      "step": 1703424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.210183619492873e-05,
+      "loss": 3.8743,
+      "step": 1703936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.209345024741821e-05,
+      "loss": 3.9089,
+      "step": 1704448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2085080678711422e-05,
+      "loss": 3.8895,
+      "step": 1704960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.20766947312009e-05,
+      "loss": 3.9054,
+      "step": 1705472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2068308783690382e-05,
+      "loss": 3.8836,
+      "step": 1705984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2059939214983595e-05,
+      "loss": 3.8909,
+      "step": 1706496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2051553267473075e-05,
+      "loss": 3.8809,
+      "step": 1707008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.204316731996255e-05,
+      "loss": 3.8992,
+      "step": 1707520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.203478137245203e-05,
+      "loss": 3.8942,
+      "step": 1708032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.202639542494151e-05,
+      "loss": 3.9028,
+      "step": 1708544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.201800947743099e-05,
+      "loss": 3.8998,
+      "step": 1709056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2009623529920474e-05,
+      "loss": 3.8824,
+      "step": 1709568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2001237582409954e-05,
+      "loss": 3.8868,
+      "step": 1710080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1992868013703163e-05,
+      "loss": 3.904,
+      "step": 1710592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1984482066192643e-05,
+      "loss": 3.8819,
+      "step": 1711104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1976096118682123e-05,
+      "loss": 3.8882,
+      "step": 1711616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1967710171171603e-05,
+      "loss": 3.8935,
+      "step": 1712128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1959340602464816e-05,
+      "loss": 3.9024,
+      "step": 1712640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1950971033758025e-05,
+      "loss": 3.8782,
+      "step": 1713152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1942585086247505e-05,
+      "loss": 3.8821,
+      "step": 1713664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1934199138736985e-05,
+      "loss": 3.8838,
+      "step": 1714176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1925813191226465e-05,
+      "loss": 3.8911,
+      "step": 1714688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1917443622519677e-05,
+      "loss": 3.8969,
+      "step": 1715200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1909057675009157e-05,
+      "loss": 3.8947,
+      "step": 1715712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1900671727498637e-05,
+      "loss": 3.8958,
+      "step": 1716224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1892285779988117e-05,
+      "loss": 3.885,
+      "step": 1716736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1883899832477597e-05,
+      "loss": 3.9121,
+      "step": 1717248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1875513884967077e-05,
+      "loss": 3.8807,
+      "step": 1717760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1867127937456554e-05,
+      "loss": 3.898,
+      "step": 1718272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1858741989946037e-05,
+      "loss": 3.8946,
+      "step": 1718784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1850356042435517e-05,
+      "loss": 3.8911,
+      "step": 1719296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1841986473728726e-05,
+      "loss": 3.8976,
+      "step": 1719808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1833600526218206e-05,
+      "loss": 3.8967,
+      "step": 1720320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.182523095751142e-05,
+      "loss": 3.8916,
+      "step": 1720832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.18168450100009e-05,
+      "loss": 3.8988,
+      "step": 1721344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1808459062490378e-05,
+      "loss": 3.8822,
+      "step": 1721856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.180008949378359e-05,
+      "loss": 3.8951,
+      "step": 1722368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.179170354627307e-05,
+      "loss": 3.8879,
+      "step": 1722880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.178331759876255e-05,
+      "loss": 3.8986,
+      "step": 1723392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1774931651252027e-05,
+      "loss": 3.8902,
+      "step": 1723904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1766545703741507e-05,
+      "loss": 3.884,
+      "step": 1724416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.175815975623099e-05,
+      "loss": 3.8905,
+      "step": 1724928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.174977380872047e-05,
+      "loss": 3.895,
+      "step": 1725440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.174138786120995e-05,
+      "loss": 3.8921,
+      "step": 1725952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.173300191369943e-05,
+      "loss": 3.8795,
+      "step": 1726464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1724615966188907e-05,
+      "loss": 3.8971,
+      "step": 1726976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1716230018678387e-05,
+      "loss": 3.9006,
+      "step": 1727488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1707844071167867e-05,
+      "loss": 3.8926,
+      "step": 1728000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1699474502461083e-05,
+      "loss": 3.8739,
+      "step": 1728512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.169108855495056e-05,
+      "loss": 3.8909,
+      "step": 1729024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.168270260744004e-05,
+      "loss": 3.8857,
+      "step": 1729536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.167431665992952e-05,
+      "loss": 3.8984,
+      "step": 1730048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1665947091222732e-05,
+      "loss": 3.8978,
+      "step": 1730560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.165756114371221e-05,
+      "loss": 3.8847,
+      "step": 1731072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.164917519620169e-05,
+      "loss": 3.8858,
+      "step": 1731584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.164078924869117e-05,
+      "loss": 3.8972,
+      "step": 1732096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.163241967998438e-05,
+      "loss": 3.8758,
+      "step": 1732608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.162403373247386e-05,
+      "loss": 3.8905,
+      "step": 1733120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1615664163767073e-05,
+      "loss": 3.8833,
+      "step": 1733632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1607278216256553e-05,
+      "loss": 3.8988,
+      "step": 1734144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1598908647549766e-05,
+      "loss": 3.8741,
+      "step": 1734656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1590522700039246e-05,
+      "loss": 3.8826,
+      "step": 1735168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1582136752528726e-05,
+      "loss": 3.8838,
+      "step": 1735680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1573750805018205e-05,
+      "loss": 3.8945,
+      "step": 1736192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1565364857507682e-05,
+      "loss": 3.8851,
+      "step": 1736704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1556978909997162e-05,
+      "loss": 3.8815,
+      "step": 1737216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1548592962486645e-05,
+      "loss": 3.8928,
+      "step": 1737728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1540207014976125e-05,
+      "loss": 3.8955,
+      "step": 1738240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1531837446269334e-05,
+      "loss": 3.8822,
+      "step": 1738752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1523451498758814e-05,
+      "loss": 3.8974,
+      "step": 1739264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1515065551248294e-05,
+      "loss": 3.885,
+      "step": 1739776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1506679603737774e-05,
+      "loss": 3.8912,
+      "step": 1740288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1498293656227254e-05,
+      "loss": 3.883,
+      "step": 1740800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1489924087520467e-05,
+      "loss": 3.891,
+      "step": 1741312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1481538140009947e-05,
+      "loss": 3.8802,
+      "step": 1741824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1473152192499427e-05,
+      "loss": 3.8891,
+      "step": 1742336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1464766244988907e-05,
+      "loss": 3.8912,
+      "step": 1742848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1456380297478387e-05,
+      "loss": 3.8889,
+      "step": 1743360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.14480107287716e-05,
+      "loss": 3.8833,
+      "step": 1743872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1439641160064808e-05,
+      "loss": 3.8902,
+      "step": 1744384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1431255212554288e-05,
+      "loss": 3.8727,
+      "step": 1744896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1422869265043768e-05,
+      "loss": 3.9037,
+      "step": 1745408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1414483317533248e-05,
+      "loss": 3.8886,
+      "step": 1745920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1406097370022728e-05,
+      "loss": 3.894,
+      "step": 1746432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1397711422512208e-05,
+      "loss": 3.8833,
+      "step": 1746944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1389325475001688e-05,
+      "loss": 3.9028,
+      "step": 1747456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1380939527491168e-05,
+      "loss": 3.8924,
+      "step": 1747968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1372553579980648e-05,
+      "loss": 3.8851,
+      "step": 1748480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1364167632470128e-05,
+      "loss": 3.8874,
+      "step": 1748992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1355781684959608e-05,
+      "loss": 3.8913,
+      "step": 1749504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1347395737449088e-05,
+      "loss": 3.8931,
+      "step": 1750016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.13390261687423e-05,
+      "loss": 3.9028,
+      "step": 1750528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.133064022123178e-05,
+      "loss": 3.8755,
+      "step": 1751040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.132227065252499e-05,
+      "loss": 3.8944,
+      "step": 1751552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.131388470501447e-05,
+      "loss": 3.8866,
+      "step": 1752064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.130549875750395e-05,
+      "loss": 3.8972,
+      "step": 1752576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.129711280999343e-05,
+      "loss": 3.89,
+      "step": 1753088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.128872686248291e-05,
+      "loss": 3.8838,
+      "step": 1753600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1280340914972392e-05,
+      "loss": 3.8973,
+      "step": 1754112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.127195496746187e-05,
+      "loss": 3.8888,
+      "step": 1754624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.126356901995135e-05,
+      "loss": 3.8939,
+      "step": 1755136
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.043178081512451,
+      "eval_runtime": 297.8954,
+      "eval_samples_per_second": 1280.956,
+      "eval_steps_per_second": 40.031,
+      "step": 1755360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.125519945124456e-05,
+      "loss": 3.8882,
+      "step": 1755648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1246813503734038e-05,
+      "loss": 3.8883,
+      "step": 1756160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.123842755622352e-05,
+      "loss": 3.8943,
+      "step": 1756672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1230041608713e-05,
+      "loss": 3.8935,
+      "step": 1757184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.122165566120248e-05,
+      "loss": 3.9007,
+      "step": 1757696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.121326971369196e-05,
+      "loss": 3.8878,
+      "step": 1758208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.120488376618144e-05,
+      "loss": 3.8803,
+      "step": 1758720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.119649781867092e-05,
+      "loss": 3.8805,
+      "step": 1759232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1188111871160398e-05,
+      "loss": 3.8951,
+      "step": 1759744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1179725923649878e-05,
+      "loss": 3.8985,
+      "step": 1760256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.117133997613936e-05,
+      "loss": 3.8838,
+      "step": 1760768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.116295402862884e-05,
+      "loss": 3.8938,
+      "step": 1761280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.115458445992205e-05,
+      "loss": 3.8856,
+      "step": 1761792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.114619851241153e-05,
+      "loss": 3.8846,
+      "step": 1762304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.113781256490101e-05,
+      "loss": 3.8822,
+      "step": 1762816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.112942661739049e-05,
+      "loss": 3.8799,
+      "step": 1763328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.112104066987997e-05,
+      "loss": 3.8851,
+      "step": 1763840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1112654722369453e-05,
+      "loss": 3.882,
+      "step": 1764352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.110426877485893e-05,
+      "loss": 3.8801,
+      "step": 1764864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.109588282734841e-05,
+      "loss": 3.913,
+      "step": 1765376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.108749687983789e-05,
+      "loss": 3.8894,
+      "step": 1765888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.107911093232737e-05,
+      "loss": 3.8875,
+      "step": 1766400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.107072498481685e-05,
+      "loss": 3.8927,
+      "step": 1766912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.106233903730633e-05,
+      "loss": 3.8951,
+      "step": 1767424
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1053985847403274e-05,
+      "loss": 3.8766,
+      "step": 1767936
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1045599899892754e-05,
+      "loss": 3.8883,
+      "step": 1768448
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.103721395238223e-05,
+      "loss": 3.8866,
+      "step": 1768960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.102882800487171e-05,
+      "loss": 3.8818,
+      "step": 1769472
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.102044205736119e-05,
+      "loss": 3.8792,
+      "step": 1769984
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.101205610985067e-05,
+      "loss": 3.8787,
+      "step": 1770496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1003686541143883e-05,
+      "loss": 3.8884,
+      "step": 1771008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0995300593633363e-05,
+      "loss": 3.8941,
+      "step": 1771520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0986914646122843e-05,
+      "loss": 3.892,
+      "step": 1772032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0978528698612323e-05,
+      "loss": 3.8834,
+      "step": 1772544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0970159129905532e-05,
+      "loss": 3.8912,
+      "step": 1773056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0961773182395016e-05,
+      "loss": 3.8889,
+      "step": 1773568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0953387234884496e-05,
+      "loss": 3.885,
+      "step": 1774080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0945001287373976e-05,
+      "loss": 3.8917,
+      "step": 1774592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0936615339863455e-05,
+      "loss": 3.8752,
+      "step": 1775104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0928229392352935e-05,
+      "loss": 3.8913,
+      "step": 1775616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0919859823646145e-05,
+      "loss": 3.8761,
+      "step": 1776128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0911473876135625e-05,
+      "loss": 3.8997,
+      "step": 1776640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0903087928625108e-05,
+      "loss": 3.89,
+      "step": 1777152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0894701981114584e-05,
+      "loss": 3.8889,
+      "step": 1777664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0886316033604064e-05,
+      "loss": 3.8942,
+      "step": 1778176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0877930086093544e-05,
+      "loss": 3.8839,
+      "step": 1778688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0869544138583024e-05,
+      "loss": 3.8919,
+      "step": 1779200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0861158191072504e-05,
+      "loss": 3.8832,
+      "step": 1779712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0852788622365717e-05,
+      "loss": 3.864,
+      "step": 1780224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0844402674855197e-05,
+      "loss": 3.9041,
+      "step": 1780736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0836016727344677e-05,
+      "loss": 3.8833,
+      "step": 1781248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0827630779834157e-05,
+      "loss": 3.8983,
+      "step": 1781760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0819244832323637e-05,
+      "loss": 3.8799,
+      "step": 1782272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0810858884813113e-05,
+      "loss": 3.8836,
+      "step": 1782784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0802472937302593e-05,
+      "loss": 3.8722,
+      "step": 1783296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0794086989792076e-05,
+      "loss": 3.8934,
+      "step": 1783808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0785701042281556e-05,
+      "loss": 3.8882,
+      "step": 1784320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0777315094771036e-05,
+      "loss": 3.8933,
+      "step": 1784832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0768929147260516e-05,
+      "loss": 3.8995,
+      "step": 1785344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0760543199749996e-05,
+      "loss": 3.8794,
+      "step": 1785856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0752173631043205e-05,
+      "loss": 3.8818,
+      "step": 1786368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0743787683532685e-05,
+      "loss": 3.898,
+      "step": 1786880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0735418114825898e-05,
+      "loss": 3.8768,
+      "step": 1787392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0727032167315378e-05,
+      "loss": 3.8787,
+      "step": 1787904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0718646219804858e-05,
+      "loss": 3.8871,
+      "step": 1788416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0710260272294338e-05,
+      "loss": 3.892,
+      "step": 1788928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0701874324783818e-05,
+      "loss": 3.8754,
+      "step": 1789440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.069350475607703e-05,
+      "loss": 3.8776,
+      "step": 1789952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.068511880856651e-05,
+      "loss": 3.8715,
+      "step": 1790464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.067673286105599e-05,
+      "loss": 3.8847,
+      "step": 1790976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.066834691354547e-05,
+      "loss": 3.8912,
+      "step": 1791488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0659960966034947e-05,
+      "loss": 3.8887,
+      "step": 1792000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.065159139732816e-05,
+      "loss": 3.8843,
+      "step": 1792512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.064320544981764e-05,
+      "loss": 3.8843,
+      "step": 1793024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.063481950230712e-05,
+      "loss": 3.9038,
+      "step": 1793536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.06264335547966e-05,
+      "loss": 3.8761,
+      "step": 1794048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.061804760728608e-05,
+      "loss": 3.8962,
+      "step": 1794560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.060966165977556e-05,
+      "loss": 3.8871,
+      "step": 1795072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.060127571226504e-05,
+      "loss": 3.8815,
+      "step": 1795584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.059288976475452e-05,
+      "loss": 3.8942,
+      "step": 1796096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.058452019604773e-05,
+      "loss": 3.8899,
+      "step": 1796608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0576150627340944e-05,
+      "loss": 3.8856,
+      "step": 1797120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.056776467983042e-05,
+      "loss": 3.8927,
+      "step": 1797632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.05593787323199e-05,
+      "loss": 3.8777,
+      "step": 1798144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.055099278480938e-05,
+      "loss": 3.8876,
+      "step": 1798656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.054260683729886e-05,
+      "loss": 3.8839,
+      "step": 1799168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.053422088978834e-05,
+      "loss": 3.895,
+      "step": 1799680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0525834942277823e-05,
+      "loss": 3.882,
+      "step": 1800192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0517465373571032e-05,
+      "loss": 3.8821,
+      "step": 1800704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.050909580486424e-05,
+      "loss": 3.8784,
+      "step": 1801216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.050070985735372e-05,
+      "loss": 3.8892,
+      "step": 1801728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.04923239098432e-05,
+      "loss": 3.887,
+      "step": 1802240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0483937962332685e-05,
+      "loss": 3.8721,
+      "step": 1802752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0475552014822165e-05,
+      "loss": 3.8915,
+      "step": 1803264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0467166067311645e-05,
+      "loss": 3.8916,
+      "step": 1803776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0458780119801125e-05,
+      "loss": 3.8856,
+      "step": 1804288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.04503941722906e-05,
+      "loss": 3.8674,
+      "step": 1804800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.044200822478008e-05,
+      "loss": 3.8835,
+      "step": 1805312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.043362227726956e-05,
+      "loss": 3.8785,
+      "step": 1805824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.042523632975904e-05,
+      "loss": 3.8954,
+      "step": 1806336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0416866761052254e-05,
+      "loss": 3.892,
+      "step": 1806848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0408480813541734e-05,
+      "loss": 3.8808,
+      "step": 1807360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0400094866031214e-05,
+      "loss": 3.8819,
+      "step": 1807872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0391708918520693e-05,
+      "loss": 3.8852,
+      "step": 1808384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0383339349813903e-05,
+      "loss": 3.8755,
+      "step": 1808896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.037496978110712e-05,
+      "loss": 3.8816,
+      "step": 1809408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.03665838335966e-05,
+      "loss": 3.8768,
+      "step": 1809920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0358197886086075e-05,
+      "loss": 3.8965,
+      "step": 1810432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0349811938575555e-05,
+      "loss": 3.8685,
+      "step": 1810944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0341425991065035e-05,
+      "loss": 3.878,
+      "step": 1811456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0333056422358247e-05,
+      "loss": 3.8763,
+      "step": 1811968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.032468685365146e-05,
+      "loss": 3.8887,
+      "step": 1812480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.031630090614094e-05,
+      "loss": 3.8755,
+      "step": 1812992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.030791495863042e-05,
+      "loss": 3.8753,
+      "step": 1813504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0299529011119896e-05,
+      "loss": 3.8885,
+      "step": 1814016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0291143063609376e-05,
+      "loss": 3.8918,
+      "step": 1814528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0282757116098856e-05,
+      "loss": 3.8745,
+      "step": 1815040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.027437116858834e-05,
+      "loss": 3.89,
+      "step": 1815552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.026598522107782e-05,
+      "loss": 3.879,
+      "step": 1816064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.02575992735673e-05,
+      "loss": 3.8855,
+      "step": 1816576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.024921332605678e-05,
+      "loss": 3.8791,
+      "step": 1817088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0240827378546256e-05,
+      "loss": 3.884,
+      "step": 1817600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0232441431035736e-05,
+      "loss": 3.8764,
+      "step": 1818112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0224055483525216e-05,
+      "loss": 3.8844,
+      "step": 1818624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.021568591481843e-05,
+      "loss": 3.8877,
+      "step": 1819136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.020729996730791e-05,
+      "loss": 3.8832,
+      "step": 1819648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.019891401979739e-05,
+      "loss": 3.8791,
+      "step": 1820160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.01905444510906e-05,
+      "loss": 3.8827,
+      "step": 1820672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0182158503580077e-05,
+      "loss": 3.8653,
+      "step": 1821184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0173772556069557e-05,
+      "loss": 3.8996,
+      "step": 1821696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.016538660855904e-05,
+      "loss": 3.8795,
+      "step": 1822208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.015700066104852e-05,
+      "loss": 3.8867,
+      "step": 1822720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0148614713538e-05,
+      "loss": 3.8765,
+      "step": 1823232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.014022876602748e-05,
+      "loss": 3.8949,
+      "step": 1823744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.013184281851696e-05,
+      "loss": 3.8879,
+      "step": 1824256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.012347324981017e-05,
+      "loss": 3.8806,
+      "step": 1824768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.011508730229965e-05,
+      "loss": 3.8846,
+      "step": 1825280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0106701354789133e-05,
+      "loss": 3.8832,
+      "step": 1825792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.009831540727861e-05,
+      "loss": 3.8878,
+      "step": 1826304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0089945838571822e-05,
+      "loss": 3.9006,
+      "step": 1826816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.008157626986503e-05,
+      "loss": 3.8667,
+      "step": 1827328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.007319032235451e-05,
+      "loss": 3.8925,
+      "step": 1827840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0064804374843994e-05,
+      "loss": 3.8802,
+      "step": 1828352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0056418427333474e-05,
+      "loss": 3.8887,
+      "step": 1828864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0048048858626683e-05,
+      "loss": 3.8785,
+      "step": 1829376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0039662911116163e-05,
+      "loss": 3.8824,
+      "step": 1829888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0031276963605643e-05,
+      "loss": 3.8893,
+      "step": 1830400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0022891016095123e-05,
+      "loss": 3.8841,
+      "step": 1830912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0014505068584603e-05,
+      "loss": 3.8858,
+      "step": 1831424
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.039677619934082,
+      "eval_runtime": 296.203,
+      "eval_samples_per_second": 1288.275,
+      "eval_steps_per_second": 40.26,
+      "step": 1831680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.0006119121074083e-05,
+      "loss": 3.8845,
+      "step": 1831936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9997733173563563e-05,
+      "loss": 3.8811,
+      "step": 1832448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9989347226053043e-05,
+      "loss": 3.8889,
+      "step": 1832960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9980977657346256e-05,
+      "loss": 3.8847,
+      "step": 1833472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9972608088639465e-05,
+      "loss": 3.895,
+      "step": 1833984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9964222141128948e-05,
+      "loss": 3.8785,
+      "step": 1834496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9955836193618428e-05,
+      "loss": 3.8779,
+      "step": 1835008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9947450246107908e-05,
+      "loss": 3.8779,
+      "step": 1835520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9939064298597384e-05,
+      "loss": 3.8845,
+      "step": 1836032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9930678351086864e-05,
+      "loss": 3.8917,
+      "step": 1836544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9922292403576344e-05,
+      "loss": 3.8791,
+      "step": 1837056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9913922834869557e-05,
+      "loss": 3.8871,
+      "step": 1837568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.990555326616277e-05,
+      "loss": 3.8839,
+      "step": 1838080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.989716731865225e-05,
+      "loss": 3.8767,
+      "step": 1838592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.988878137114173e-05,
+      "loss": 3.8814,
+      "step": 1839104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9880395423631206e-05,
+      "loss": 3.8719,
+      "step": 1839616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9872009476120686e-05,
+      "loss": 3.8803,
+      "step": 1840128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.98636399074139e-05,
+      "loss": 3.8726,
+      "step": 1840640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.985525395990338e-05,
+      "loss": 3.8757,
+      "step": 1841152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9846868012392858e-05,
+      "loss": 3.9027,
+      "step": 1841664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9838482064882338e-05,
+      "loss": 3.8889,
+      "step": 1842176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9830096117371818e-05,
+      "loss": 3.8813,
+      "step": 1842688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9821710169861298e-05,
+      "loss": 3.8853,
+      "step": 1843200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.981334060115451e-05,
+      "loss": 3.8907,
+      "step": 1843712
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.980495465364399e-05,
+      "loss": 3.8665,
+      "step": 1844224
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.979656870613347e-05,
+      "loss": 3.8836,
+      "step": 1844736
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.978818275862295e-05,
+      "loss": 3.8816,
+      "step": 1845248
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.977979681111243e-05,
+      "loss": 3.8732,
+      "step": 1845760
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.977141086360191e-05,
+      "loss": 3.8706,
+      "step": 1846272
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9763024916091387e-05,
+      "loss": 3.8764,
+      "step": 1846784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.975463896858087e-05,
+      "loss": 3.8809,
+      "step": 1847296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9746269399874083e-05,
+      "loss": 3.8889,
+      "step": 1847808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9737883452363563e-05,
+      "loss": 3.8835,
+      "step": 1848320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.972949750485304e-05,
+      "loss": 3.88,
+      "step": 1848832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.972114431494998e-05,
+      "loss": 3.8853,
+      "step": 1849344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9712758367439464e-05,
+      "loss": 3.884,
+      "step": 1849856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9704372419928944e-05,
+      "loss": 3.8803,
+      "step": 1850368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9695986472418424e-05,
+      "loss": 3.885,
+      "step": 1850880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9687600524907904e-05,
+      "loss": 3.8695,
+      "step": 1851392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9679214577397384e-05,
+      "loss": 3.882,
+      "step": 1851904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.967082862988686e-05,
+      "loss": 3.8717,
+      "step": 1852416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.966244268237634e-05,
+      "loss": 3.8931,
+      "step": 1852928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.965405673486582e-05,
+      "loss": 3.8872,
+      "step": 1853440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9645670787355304e-05,
+      "loss": 3.8814,
+      "step": 1853952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9637284839844784e-05,
+      "loss": 3.889,
+      "step": 1854464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9628898892334264e-05,
+      "loss": 3.8772,
+      "step": 1854976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.962051294482374e-05,
+      "loss": 3.8903,
+      "step": 1855488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9612143376116953e-05,
+      "loss": 3.8784,
+      "step": 1856000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9603757428606433e-05,
+      "loss": 3.8533,
+      "step": 1856512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9595387859899645e-05,
+      "loss": 3.8974,
+      "step": 1857024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9587001912389125e-05,
+      "loss": 3.8772,
+      "step": 1857536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9578615964878605e-05,
+      "loss": 3.8962,
+      "step": 1858048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9570246396171814e-05,
+      "loss": 3.874,
+      "step": 1858560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9561860448661294e-05,
+      "loss": 3.8756,
+      "step": 1859072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9553474501150774e-05,
+      "loss": 3.8689,
+      "step": 1859584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9545088553640258e-05,
+      "loss": 3.8857,
+      "step": 1860096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9536702606129737e-05,
+      "loss": 3.8868,
+      "step": 1860608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9528316658619214e-05,
+      "loss": 3.8842,
+      "step": 1861120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9519930711108694e-05,
+      "loss": 3.8943,
+      "step": 1861632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9511544763598174e-05,
+      "loss": 3.8761,
+      "step": 1862144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9503175194891386e-05,
+      "loss": 3.8719,
+      "step": 1862656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9494789247380866e-05,
+      "loss": 3.8933,
+      "step": 1863168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9486403299870346e-05,
+      "loss": 3.8714,
+      "step": 1863680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9478017352359826e-05,
+      "loss": 3.877,
+      "step": 1864192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9469631404849306e-05,
+      "loss": 3.8799,
+      "step": 1864704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9461245457338786e-05,
+      "loss": 3.8897,
+      "step": 1865216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9452859509828266e-05,
+      "loss": 3.8686,
+      "step": 1865728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9444473562317746e-05,
+      "loss": 3.8753,
+      "step": 1866240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.943610399361096e-05,
+      "loss": 3.8679,
+      "step": 1866752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.942771804610044e-05,
+      "loss": 3.8769,
+      "step": 1867264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9419348477393648e-05,
+      "loss": 3.8869,
+      "step": 1867776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9410962529883128e-05,
+      "loss": 3.8805,
+      "step": 1868288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9402576582372608e-05,
+      "loss": 3.8811,
+      "step": 1868800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9394190634862088e-05,
+      "loss": 3.88,
+      "step": 1869312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9385804687351568e-05,
+      "loss": 3.9036,
+      "step": 1869824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9377418739841047e-05,
+      "loss": 3.8704,
+      "step": 1870336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9369032792330527e-05,
+      "loss": 3.8871,
+      "step": 1870848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.936066322362374e-05,
+      "loss": 3.8855,
+      "step": 1871360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.935227727611322e-05,
+      "loss": 3.8772,
+      "step": 1871872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9343891328602696e-05,
+      "loss": 3.8871,
+      "step": 1872384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.933550538109218e-05,
+      "loss": 3.8854,
+      "step": 1872896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.932711943358166e-05,
+      "loss": 3.8836,
+      "step": 1873408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.931873348607114e-05,
+      "loss": 3.8858,
+      "step": 1873920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.931034753856062e-05,
+      "loss": 3.8739,
+      "step": 1874432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.93019615910501e-05,
+      "loss": 3.8766,
+      "step": 1874944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.929359202234331e-05,
+      "loss": 3.8802,
+      "step": 1875456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.928520607483279e-05,
+      "loss": 3.886,
+      "step": 1875968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9276836506126e-05,
+      "loss": 3.8761,
+      "step": 1876480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.926845055861548e-05,
+      "loss": 3.8786,
+      "step": 1876992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.926006461110496e-05,
+      "loss": 3.8733,
+      "step": 1877504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.925167866359444e-05,
+      "loss": 3.8869,
+      "step": 1878016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.924329271608392e-05,
+      "loss": 3.8799,
+      "step": 1878528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9234923147377133e-05,
+      "loss": 3.8688,
+      "step": 1879040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9226537199866613e-05,
+      "loss": 3.8844,
+      "step": 1879552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9218151252356093e-05,
+      "loss": 3.8867,
+      "step": 1880064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9209765304845573e-05,
+      "loss": 3.8808,
+      "step": 1880576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.920137935733505e-05,
+      "loss": 3.8627,
+      "step": 1881088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9193009788628262e-05,
+      "loss": 3.877,
+      "step": 1881600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9184623841117742e-05,
+      "loss": 3.8708,
+      "step": 1882112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9176237893607226e-05,
+      "loss": 3.889,
+      "step": 1882624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9167851946096702e-05,
+      "loss": 3.8839,
+      "step": 1883136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9159482377389915e-05,
+      "loss": 3.8837,
+      "step": 1883648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9151096429879395e-05,
+      "loss": 3.8746,
+      "step": 1884160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9142710482368875e-05,
+      "loss": 3.8766,
+      "step": 1884672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.913432453485835e-05,
+      "loss": 3.8732,
+      "step": 1885184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9125938587347835e-05,
+      "loss": 3.8762,
+      "step": 1885696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9117569018641047e-05,
+      "loss": 3.8701,
+      "step": 1886208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9109183071130524e-05,
+      "loss": 3.8885,
+      "step": 1886720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9100797123620004e-05,
+      "loss": 3.8675,
+      "step": 1887232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9092411176109483e-05,
+      "loss": 3.8686,
+      "step": 1887744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9084025228598963e-05,
+      "loss": 3.8736,
+      "step": 1888256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9075655659892176e-05,
+      "loss": 3.8747,
+      "step": 1888768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9067269712381656e-05,
+      "loss": 3.8751,
+      "step": 1889280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.905890014367487e-05,
+      "loss": 3.8709,
+      "step": 1889792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.905051419616435e-05,
+      "loss": 3.8801,
+      "step": 1890304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9042128248653825e-05,
+      "loss": 3.8881,
+      "step": 1890816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9033742301143305e-05,
+      "loss": 3.8675,
+      "step": 1891328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9025356353632788e-05,
+      "loss": 3.8845,
+      "step": 1891840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9016970406122268e-05,
+      "loss": 3.8751,
+      "step": 1892352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9008584458611748e-05,
+      "loss": 3.8823,
+      "step": 1892864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9000198511101228e-05,
+      "loss": 3.8763,
+      "step": 1893376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8991812563590705e-05,
+      "loss": 3.8751,
+      "step": 1893888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8983426616080185e-05,
+      "loss": 3.8737,
+      "step": 1894400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8975040668569665e-05,
+      "loss": 3.8792,
+      "step": 1894912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8966654721059144e-05,
+      "loss": 3.8801,
+      "step": 1895424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8958285152352357e-05,
+      "loss": 3.8756,
+      "step": 1895936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8949899204841837e-05,
+      "loss": 3.8744,
+      "step": 1896448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.894152963613505e-05,
+      "loss": 3.8755,
+      "step": 1896960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8933143688624526e-05,
+      "loss": 3.8627,
+      "step": 1897472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8924757741114006e-05,
+      "loss": 3.8926,
+      "step": 1897984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.891637179360349e-05,
+      "loss": 3.8719,
+      "step": 1898496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.890798584609297e-05,
+      "loss": 3.8866,
+      "step": 1899008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.889959989858245e-05,
+      "loss": 3.8691,
+      "step": 1899520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8891230329875658e-05,
+      "loss": 3.8917,
+      "step": 1900032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8882844382365138e-05,
+      "loss": 3.882,
+      "step": 1900544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.887447481365835e-05,
+      "loss": 3.8751,
+      "step": 1901056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.886608886614783e-05,
+      "loss": 3.8786,
+      "step": 1901568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.885770291863731e-05,
+      "loss": 3.8787,
+      "step": 1902080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.884931697112679e-05,
+      "loss": 3.8824,
+      "step": 1902592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.884093102361627e-05,
+      "loss": 3.8918,
+      "step": 1903104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.883254507610575e-05,
+      "loss": 3.864,
+      "step": 1903616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.882415912859523e-05,
+      "loss": 3.8886,
+      "step": 1904128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.881577318108471e-05,
+      "loss": 3.8727,
+      "step": 1904640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8807403612377923e-05,
+      "loss": 3.884,
+      "step": 1905152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8799017664867403e-05,
+      "loss": 3.8782,
+      "step": 1905664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8790631717356883e-05,
+      "loss": 3.8749,
+      "step": 1906176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.878224576984636e-05,
+      "loss": 3.8822,
+      "step": 1906688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.877385982233584e-05,
+      "loss": 3.8829,
+      "step": 1907200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.876547387482532e-05,
+      "loss": 3.8795,
+      "step": 1907712
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.038036823272705,
+      "eval_runtime": 295.7799,
+      "eval_samples_per_second": 1290.118,
+      "eval_steps_per_second": 40.317,
+      "step": 1908000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.87570879273148e-05,
+      "loss": 3.8736,
+      "step": 1908224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8748701979804283e-05,
+      "loss": 3.8789,
+      "step": 1908736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8740316032293763e-05,
+      "loss": 3.8858,
+      "step": 1909248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8731930084783243e-05,
+      "loss": 3.8802,
+      "step": 1909760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.872354413727272e-05,
+      "loss": 3.8884,
+      "step": 1910272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.87151581897622e-05,
+      "loss": 3.8758,
+      "step": 1910784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.870677224225168e-05,
+      "loss": 3.8757,
+      "step": 1911296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.869838629474116e-05,
+      "loss": 3.875,
+      "step": 1911808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8690000347230642e-05,
+      "loss": 3.8791,
+      "step": 1912320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8681614399720122e-05,
+      "loss": 3.8851,
+      "step": 1912832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.86732284522096e-05,
+      "loss": 3.8782,
+      "step": 1913344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.866484250469908e-05,
+      "loss": 3.8814,
+      "step": 1913856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.865647293599229e-05,
+      "loss": 3.8827,
+      "step": 1914368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.864808698848177e-05,
+      "loss": 3.8703,
+      "step": 1914880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.863970104097125e-05,
+      "loss": 3.8758,
+      "step": 1915392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.863131509346073e-05,
+      "loss": 3.8683,
+      "step": 1915904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.862292914595021e-05,
+      "loss": 3.8722,
+      "step": 1916416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.861454319843969e-05,
+      "loss": 3.8705,
+      "step": 1916928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.860615725092917e-05,
+      "loss": 3.8716,
+      "step": 1917440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.859777130341865e-05,
+      "loss": 3.8938,
+      "step": 1917952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.858940173471186e-05,
+      "loss": 3.885,
+      "step": 1918464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8581015787201343e-05,
+      "loss": 3.8781,
+      "step": 1918976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8572629839690823e-05,
+      "loss": 3.8802,
+      "step": 1919488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8564243892180303e-05,
+      "loss": 3.8877,
+      "step": 1920000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.855585794466978e-05,
+      "loss": 3.8619,
+      "step": 1920512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.854747199715926e-05,
+      "loss": 3.8778,
+      "step": 1921024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.853908604964874e-05,
+      "loss": 3.8774,
+      "step": 1921536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.853070010213822e-05,
+      "loss": 3.8643,
+      "step": 1922048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8522330533431432e-05,
+      "loss": 3.8733,
+      "step": 1922560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8513944585920912e-05,
+      "loss": 3.8675,
+      "step": 1923072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8505575017214125e-05,
+      "loss": 3.8769,
+      "step": 1923584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8497189069703605e-05,
+      "loss": 3.8785,
+      "step": 1924096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.848880312219308e-05,
+      "loss": 3.8783,
+      "step": 1924608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.848041717468256e-05,
+      "loss": 3.8826,
+      "step": 1925120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8472047605975777e-05,
+      "loss": 3.8739,
+      "step": 1925632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8463661658465254e-05,
+      "loss": 3.8791,
+      "step": 1926144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8455292089758466e-05,
+      "loss": 3.8817,
+      "step": 1926656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8446906142247946e-05,
+      "loss": 3.8743,
+      "step": 1927168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8438520194737426e-05,
+      "loss": 3.8672,
+      "step": 1927680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8430134247226906e-05,
+      "loss": 3.876,
+      "step": 1928192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8421748299716386e-05,
+      "loss": 3.8688,
+      "step": 1928704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8413362352205866e-05,
+      "loss": 3.8869,
+      "step": 1929216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8404976404695346e-05,
+      "loss": 3.8839,
+      "step": 1929728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8396590457184826e-05,
+      "loss": 3.8742,
+      "step": 1930240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8388204509674306e-05,
+      "loss": 3.8836,
+      "step": 1930752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8379834940967515e-05,
+      "loss": 3.8709,
+      "step": 1931264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8371448993456998e-05,
+      "loss": 3.8812,
+      "step": 1931776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8363063045946478e-05,
+      "loss": 3.8786,
+      "step": 1932288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8354677098435958e-05,
+      "loss": 3.846,
+      "step": 1932800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8346291150925435e-05,
+      "loss": 3.8912,
+      "step": 1933312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8337905203414915e-05,
+      "loss": 3.8746,
+      "step": 1933824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8329519255904395e-05,
+      "loss": 3.8857,
+      "step": 1934336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8321133308393874e-05,
+      "loss": 3.8725,
+      "step": 1934848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8312763739687087e-05,
+      "loss": 3.8722,
+      "step": 1935360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8304377792176567e-05,
+      "loss": 3.8627,
+      "step": 1935872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.829600822346978e-05,
+      "loss": 3.8809,
+      "step": 1936384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.828763865476299e-05,
+      "loss": 3.8814,
+      "step": 1936896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.827925270725247e-05,
+      "loss": 3.877,
+      "step": 1937408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8270866759741952e-05,
+      "loss": 3.8943,
+      "step": 1937920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8262480812231432e-05,
+      "loss": 3.8724,
+      "step": 1938432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.825409486472091e-05,
+      "loss": 3.8625,
+      "step": 1938944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8245708917210388e-05,
+      "loss": 3.8916,
+      "step": 1939456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8237322969699868e-05,
+      "loss": 3.8672,
+      "step": 1939968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8228937022189348e-05,
+      "loss": 3.8644,
+      "step": 1940480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.822056745348256e-05,
+      "loss": 3.881,
+      "step": 1940992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.821218150597204e-05,
+      "loss": 3.8789,
+      "step": 1941504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.820379555846152e-05,
+      "loss": 3.8669,
+      "step": 1942016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8195409610951e-05,
+      "loss": 3.8677,
+      "step": 1942528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.818702366344048e-05,
+      "loss": 3.8625,
+      "step": 1943040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.817865409473369e-05,
+      "loss": 3.8716,
+      "step": 1943552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.817026814722317e-05,
+      "loss": 3.8839,
+      "step": 1944064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8161898578516382e-05,
+      "loss": 3.8757,
+      "step": 1944576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8153512631005862e-05,
+      "loss": 3.8774,
+      "step": 1945088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8145126683495342e-05,
+      "loss": 3.8699,
+      "step": 1945600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8136740735984822e-05,
+      "loss": 3.8944,
+      "step": 1946112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8128354788474302e-05,
+      "loss": 3.8679,
+      "step": 1946624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8119968840963782e-05,
+      "loss": 3.8841,
+      "step": 1947136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8111582893453262e-05,
+      "loss": 3.8763,
+      "step": 1947648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8103196945942742e-05,
+      "loss": 3.8736,
+      "step": 1948160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.809481099843222e-05,
+      "loss": 3.8783,
+      "step": 1948672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8086441429725434e-05,
+      "loss": 3.8822,
+      "step": 1949184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.807805548221491e-05,
+      "loss": 3.8736,
+      "step": 1949696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.806966953470439e-05,
+      "loss": 3.8816,
+      "step": 1950208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8061283587193874e-05,
+      "loss": 3.8692,
+      "step": 1950720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8052897639683354e-05,
+      "loss": 3.8736,
+      "step": 1951232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8044511692172834e-05,
+      "loss": 3.8744,
+      "step": 1951744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8036125744662314e-05,
+      "loss": 3.8803,
+      "step": 1952256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8027756175955523e-05,
+      "loss": 3.8723,
+      "step": 1952768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8019370228445003e-05,
+      "loss": 3.8746,
+      "step": 1953280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8010984280934483e-05,
+      "loss": 3.8704,
+      "step": 1953792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8002598333423963e-05,
+      "loss": 3.8778,
+      "step": 1954304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7994212385913446e-05,
+      "loss": 3.8779,
+      "step": 1954816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7985842817206655e-05,
+      "loss": 3.8636,
+      "step": 1955328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7977456869696135e-05,
+      "loss": 3.8828,
+      "step": 1955840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7969070922185615e-05,
+      "loss": 3.8777,
+      "step": 1956352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7960684974675092e-05,
+      "loss": 3.881,
+      "step": 1956864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7952299027164575e-05,
+      "loss": 3.8565,
+      "step": 1957376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7943913079654055e-05,
+      "loss": 3.8708,
+      "step": 1957888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7935527132143535e-05,
+      "loss": 3.8679,
+      "step": 1958400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7927141184633015e-05,
+      "loss": 3.8853,
+      "step": 1958912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7918771615926224e-05,
+      "loss": 3.8788,
+      "step": 1959424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7910385668415704e-05,
+      "loss": 3.8782,
+      "step": 1959936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7901999720905184e-05,
+      "loss": 3.8728,
+      "step": 1960448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7893613773394667e-05,
+      "loss": 3.8688,
+      "step": 1960960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7885244204687876e-05,
+      "loss": 3.8653,
+      "step": 1961472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7876858257177356e-05,
+      "loss": 3.8734,
+      "step": 1961984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7868488688470565e-05,
+      "loss": 3.8677,
+      "step": 1962496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7860102740960045e-05,
+      "loss": 3.8783,
+      "step": 1963008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.785171679344953e-05,
+      "loss": 3.8658,
+      "step": 1963520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.784333084593901e-05,
+      "loss": 3.8618,
+      "step": 1964032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.783494489842849e-05,
+      "loss": 3.8692,
+      "step": 1964544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.782655895091797e-05,
+      "loss": 3.8719,
+      "step": 1965056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.781817300340745e-05,
+      "loss": 3.8694,
+      "step": 1965568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7809787055896925e-05,
+      "loss": 3.8658,
+      "step": 1966080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7801401108386405e-05,
+      "loss": 3.8788,
+      "step": 1966592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7793015160875885e-05,
+      "loss": 3.8829,
+      "step": 1967104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7784645592169098e-05,
+      "loss": 3.8664,
+      "step": 1967616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7776259644658578e-05,
+      "loss": 3.8743,
+      "step": 1968128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7767873697148057e-05,
+      "loss": 3.8743,
+      "step": 1968640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7759487749637537e-05,
+      "loss": 3.8727,
+      "step": 1969152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7751101802127017e-05,
+      "loss": 3.8739,
+      "step": 1969664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7742715854616497e-05,
+      "loss": 3.8684,
+      "step": 1970176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.773434628590971e-05,
+      "loss": 3.8713,
+      "step": 1970688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.772596033839919e-05,
+      "loss": 3.8711,
+      "step": 1971200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.771757439088867e-05,
+      "loss": 3.8772,
+      "step": 1971712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.770918844337815e-05,
+      "loss": 3.8724,
+      "step": 1972224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.770080249586763e-05,
+      "loss": 3.8661,
+      "step": 1972736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7692416548357106e-05,
+      "loss": 3.8719,
+      "step": 1973248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7684046979650322e-05,
+      "loss": 3.8568,
+      "step": 1973760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7675661032139802e-05,
+      "loss": 3.8866,
+      "step": 1974272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.766727508462928e-05,
+      "loss": 3.8677,
+      "step": 1974784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.765888913711876e-05,
+      "loss": 3.8841,
+      "step": 1975296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.765050318960824e-05,
+      "loss": 3.8659,
+      "step": 1975808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.764211724209772e-05,
+      "loss": 3.8851,
+      "step": 1976320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.76337312945872e-05,
+      "loss": 3.8794,
+      "step": 1976832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.762534534707668e-05,
+      "loss": 3.8691,
+      "step": 1977344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.761697577836989e-05,
+      "loss": 3.8733,
+      "step": 1977856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.760858983085937e-05,
+      "loss": 3.8699,
+      "step": 1978368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.760020388334885e-05,
+      "loss": 3.8815,
+      "step": 1978880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.759181793583833e-05,
+      "loss": 3.8861,
+      "step": 1979392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.758344836713154e-05,
+      "loss": 3.8608,
+      "step": 1979904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7575062419621023e-05,
+      "loss": 3.8783,
+      "step": 1980416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7566676472110503e-05,
+      "loss": 3.8718,
+      "step": 1980928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7558306903403712e-05,
+      "loss": 3.8799,
+      "step": 1981440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7549920955893192e-05,
+      "loss": 3.8713,
+      "step": 1981952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7541535008382672e-05,
+      "loss": 3.8755,
+      "step": 1982464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7533149060872152e-05,
+      "loss": 3.8768,
+      "step": 1982976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7524779492165365e-05,
+      "loss": 3.8773,
+      "step": 1983488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7516393544654845e-05,
+      "loss": 3.8776,
+      "step": 1984000
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.036283016204834,
+      "eval_runtime": 295.2186,
+      "eval_samples_per_second": 1292.571,
+      "eval_steps_per_second": 40.394,
+      "step": 1984320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7508007597144325e-05,
+      "loss": 3.868,
+      "step": 1984512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7499621649633804e-05,
+      "loss": 3.8709,
+      "step": 1985024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7491235702123284e-05,
+      "loss": 3.8799,
+      "step": 1985536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7482866133416494e-05,
+      "loss": 3.8754,
+      "step": 1986048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7474480185905977e-05,
+      "loss": 3.8818,
+      "step": 1986560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7466094238395457e-05,
+      "loss": 3.8711,
+      "step": 1987072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7457724669688666e-05,
+      "loss": 3.8718,
+      "step": 1987584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7449338722178146e-05,
+      "loss": 3.8695,
+      "step": 1988096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7440952774667626e-05,
+      "loss": 3.8722,
+      "step": 1988608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7432566827157106e-05,
+      "loss": 3.8833,
+      "step": 1989120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7424180879646586e-05,
+      "loss": 3.8692,
+      "step": 1989632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7415794932136066e-05,
+      "loss": 3.8725,
+      "step": 1990144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7407425363429278e-05,
+      "loss": 3.8799,
+      "step": 1990656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7399039415918758e-05,
+      "loss": 3.8647,
+      "step": 1991168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7390653468408235e-05,
+      "loss": 3.8739,
+      "step": 1991680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7382267520897715e-05,
+      "loss": 3.8634,
+      "step": 1992192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7373881573387195e-05,
+      "loss": 3.8652,
+      "step": 1992704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7365495625876678e-05,
+      "loss": 3.8689,
+      "step": 1993216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7357109678366158e-05,
+      "loss": 3.8643,
+      "step": 1993728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7348723730855638e-05,
+      "loss": 3.8864,
+      "step": 1994240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7340337783345114e-05,
+      "loss": 3.8845,
+      "step": 1994752
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7331951835834594e-05,
+      "loss": 3.8732,
+      "step": 1995264
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7323565888324074e-05,
+      "loss": 3.8757,
+      "step": 1995776
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7315179940813554e-05,
+      "loss": 3.8794,
+      "step": 1996288
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7306810372106767e-05,
+      "loss": 3.8628,
+      "step": 1996800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7298424424596247e-05,
+      "loss": 3.871,
+      "step": 1997312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7290038477085727e-05,
+      "loss": 3.871,
+      "step": 1997824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7281652529575207e-05,
+      "loss": 3.8572,
+      "step": 1998336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7273282960868416e-05,
+      "loss": 3.8674,
+      "step": 1998848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.72648970133579e-05,
+      "loss": 3.864,
+      "step": 1999360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.725651106584738e-05,
+      "loss": 3.8756,
+      "step": 1999872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.724812511833686e-05,
+      "loss": 3.873,
+      "step": 2000384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7239755549630068e-05,
+      "loss": 3.8717,
+      "step": 2000896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7231369602119548e-05,
+      "loss": 3.8787,
+      "step": 2001408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.722300003341276e-05,
+      "loss": 3.8655,
+      "step": 2001920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.721461408590224e-05,
+      "loss": 3.8759,
+      "step": 2002432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.720622813839172e-05,
+      "loss": 3.8793,
+      "step": 2002944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.71978421908812e-05,
+      "loss": 3.8649,
+      "step": 2003456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.718947262217441e-05,
+      "loss": 3.8681,
+      "step": 2003968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.718108667466389e-05,
+      "loss": 3.8701,
+      "step": 2004480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.717270072715337e-05,
+      "loss": 3.8655,
+      "step": 2004992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.716431477964285e-05,
+      "loss": 3.8824,
+      "step": 2005504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7155928832132333e-05,
+      "loss": 3.8762,
+      "step": 2006016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7147542884621813e-05,
+      "loss": 3.8725,
+      "step": 2006528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7139156937111293e-05,
+      "loss": 3.8779,
+      "step": 2007040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.713077098960077e-05,
+      "loss": 3.8679,
+      "step": 2007552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.712238504209025e-05,
+      "loss": 3.876,
+      "step": 2008064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.711399909457973e-05,
+      "loss": 3.8757,
+      "step": 2008576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.710561314706921e-05,
+      "loss": 3.8452,
+      "step": 2009088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7097227199558692e-05,
+      "loss": 3.8835,
+      "step": 2009600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.70888576308519e-05,
+      "loss": 3.8685,
+      "step": 2010112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.708047168334138e-05,
+      "loss": 3.879,
+      "step": 2010624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.707208573583086e-05,
+      "loss": 3.8689,
+      "step": 2011136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.706369978832034e-05,
+      "loss": 3.8654,
+      "step": 2011648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7055330219613554e-05,
+      "loss": 3.8618,
+      "step": 2012160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7046960650906766e-05,
+      "loss": 3.8753,
+      "step": 2012672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7038574703396243e-05,
+      "loss": 3.8774,
+      "step": 2013184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7030188755885723e-05,
+      "loss": 3.8707,
+      "step": 2013696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7021802808375203e-05,
+      "loss": 3.8883,
+      "step": 2014208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7013416860864683e-05,
+      "loss": 3.8733,
+      "step": 2014720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7005030913354163e-05,
+      "loss": 3.8513,
+      "step": 2015232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6996661344647375e-05,
+      "loss": 3.8842,
+      "step": 2015744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6988275397136855e-05,
+      "loss": 3.8635,
+      "step": 2016256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6979889449626335e-05,
+      "loss": 3.8623,
+      "step": 2016768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6971503502115815e-05,
+      "loss": 3.876,
+      "step": 2017280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6963117554605295e-05,
+      "loss": 3.8737,
+      "step": 2017792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.695473160709477e-05,
+      "loss": 3.8615,
+      "step": 2018304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6946345659584255e-05,
+      "loss": 3.8664,
+      "step": 2018816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6937959712073735e-05,
+      "loss": 3.8584,
+      "step": 2019328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6929573764563215e-05,
+      "loss": 3.8669,
+      "step": 2019840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6921204195856424e-05,
+      "loss": 3.8784,
+      "step": 2020352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6912818248345904e-05,
+      "loss": 3.8737,
+      "step": 2020864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6904432300835384e-05,
+      "loss": 3.8702,
+      "step": 2021376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6896046353324864e-05,
+      "loss": 3.8653,
+      "step": 2021888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.688769316342181e-05,
+      "loss": 3.8873,
+      "step": 2022400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.687930721591129e-05,
+      "loss": 3.8682,
+      "step": 2022912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.687092126840077e-05,
+      "loss": 3.8809,
+      "step": 2023424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6862535320890245e-05,
+      "loss": 3.8675,
+      "step": 2023936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6854149373379725e-05,
+      "loss": 3.8713,
+      "step": 2024448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.684577980467294e-05,
+      "loss": 3.8774,
+      "step": 2024960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.683739385716242e-05,
+      "loss": 3.8786,
+      "step": 2025472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6829007909651898e-05,
+      "loss": 3.868,
+      "step": 2025984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6820621962141378e-05,
+      "loss": 3.8815,
+      "step": 2026496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6812236014630858e-05,
+      "loss": 3.8573,
+      "step": 2027008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6803850067120338e-05,
+      "loss": 3.8681,
+      "step": 2027520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.6795464119609817e-05,
+      "loss": 3.8725,
+      "step": 2028032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.67870781720993e-05,
+      "loss": 3.8764,
+      "step": 2028544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.677870860339251e-05,
+      "loss": 3.8677,
+      "step": 2029056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.677033903468572e-05,
+      "loss": 3.8682,
+      "step": 2029568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.67619530871752e-05,
+      "loss": 3.8666,
+      "step": 2030080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.675356713966468e-05,
+      "loss": 3.8697,
+      "step": 2030592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6745181192154162e-05,
+      "loss": 3.8741,
+      "step": 2031104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6736795244643642e-05,
+      "loss": 3.8613,
+      "step": 2031616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6728409297133122e-05,
+      "loss": 3.8736,
+      "step": 2032128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.672003972842633e-05,
+      "loss": 3.8761,
+      "step": 2032640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.671165378091581e-05,
+      "loss": 3.8766,
+      "step": 2033152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.670326783340529e-05,
+      "loss": 3.8502,
+      "step": 2033664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.669488188589477e-05,
+      "loss": 3.8678,
+      "step": 2034176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6686495938384254e-05,
+      "loss": 3.8653,
+      "step": 2034688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6678126369677464e-05,
+      "loss": 3.8773,
+      "step": 2035200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6669740422166944e-05,
+      "loss": 3.8765,
+      "step": 2035712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6661354474656424e-05,
+      "loss": 3.8709,
+      "step": 2036224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.66529685271459e-05,
+      "loss": 3.8652,
+      "step": 2036736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.664458257963538e-05,
+      "loss": 3.8687,
+      "step": 2037248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6636196632124863e-05,
+      "loss": 3.8631,
+      "step": 2037760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6627810684614343e-05,
+      "loss": 3.8692,
+      "step": 2038272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6619441115907552e-05,
+      "loss": 3.8626,
+      "step": 2038784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6611055168397032e-05,
+      "loss": 3.8721,
+      "step": 2039296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6602669220886512e-05,
+      "loss": 3.8617,
+      "step": 2039808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6594283273375992e-05,
+      "loss": 3.8586,
+      "step": 2040320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6585897325865472e-05,
+      "loss": 3.8603,
+      "step": 2040832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6577511378354956e-05,
+      "loss": 3.8648,
+      "step": 2041344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6569125430844432e-05,
+      "loss": 3.869,
+      "step": 2041856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6560739483333912e-05,
+      "loss": 3.8596,
+      "step": 2042368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6552369914627125e-05,
+      "loss": 3.8702,
+      "step": 2042880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6543983967116605e-05,
+      "loss": 3.873,
+      "step": 2043392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6535614398409817e-05,
+      "loss": 3.8642,
+      "step": 2043904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6527228450899297e-05,
+      "loss": 3.8734,
+      "step": 2044416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6518842503388777e-05,
+      "loss": 3.8676,
+      "step": 2044928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6510456555878257e-05,
+      "loss": 3.8714,
+      "step": 2045440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6502070608367733e-05,
+      "loss": 3.8675,
+      "step": 2045952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6493684660857213e-05,
+      "loss": 3.8629,
+      "step": 2046464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6485315092150426e-05,
+      "loss": 3.8702,
+      "step": 2046976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6476929144639906e-05,
+      "loss": 3.8634,
+      "step": 2047488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6468543197129386e-05,
+      "loss": 3.8734,
+      "step": 2048000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6460157249618866e-05,
+      "loss": 3.8703,
+      "step": 2048512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6451771302108346e-05,
+      "loss": 3.86,
+      "step": 2049024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6443401733401555e-05,
+      "loss": 3.8656,
+      "step": 2049536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.643503216469477e-05,
+      "loss": 3.8552,
+      "step": 2050048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.642664621718425e-05,
+      "loss": 3.8802,
+      "step": 2050560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.641826026967373e-05,
+      "loss": 3.8643,
+      "step": 2051072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6409874322163207e-05,
+      "loss": 3.8787,
+      "step": 2051584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6401488374652687e-05,
+      "loss": 3.8647,
+      "step": 2052096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6393102427142167e-05,
+      "loss": 3.8766,
+      "step": 2052608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6384716479631647e-05,
+      "loss": 3.8755,
+      "step": 2053120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6376330532121127e-05,
+      "loss": 3.8626,
+      "step": 2053632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.636794458461061e-05,
+      "loss": 3.8703,
+      "step": 2054144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6359558637100087e-05,
+      "loss": 3.8626,
+      "step": 2054656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.63511890683933e-05,
+      "loss": 3.8791,
+      "step": 2055168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.634280312088278e-05,
+      "loss": 3.8759,
+      "step": 2055680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.633441717337226e-05,
+      "loss": 3.8636,
+      "step": 2056192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.632603122586174e-05,
+      "loss": 3.8687,
+      "step": 2056704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6317661657154952e-05,
+      "loss": 3.8701,
+      "step": 2057216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6309275709644432e-05,
+      "loss": 3.8768,
+      "step": 2057728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.630088976213391e-05,
+      "loss": 3.8638,
+      "step": 2058240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6292503814623388e-05,
+      "loss": 3.8751,
+      "step": 2058752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6284117867112868e-05,
+      "loss": 3.8645,
+      "step": 2059264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.627574829840608e-05,
+      "loss": 3.8751,
+      "step": 2059776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.626736235089556e-05,
+      "loss": 3.8692,
+      "step": 2060288
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.035274982452393,
+      "eval_runtime": 294.7629,
+      "eval_samples_per_second": 1294.569,
+      "eval_steps_per_second": 40.456,
+      "step": 2060640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.625897640338504e-05,
+      "loss": 3.8615,
+      "step": 2060800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.625059045587452e-05,
+      "loss": 3.8654,
+      "step": 2061312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6242204508364e-05,
+      "loss": 3.8725,
+      "step": 2061824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.623381856085348e-05,
+      "loss": 3.8744,
+      "step": 2062336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.622543261334296e-05,
+      "loss": 3.8787,
+      "step": 2062848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.621704666583244e-05,
+      "loss": 3.8653,
+      "step": 2063360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.620866071832192e-05,
+      "loss": 3.8676,
+      "step": 2063872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.62002747708114e-05,
+      "loss": 3.8668,
+      "step": 2064384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.619188882330088e-05,
+      "loss": 3.8703,
+      "step": 2064896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.618350287579036e-05,
+      "loss": 3.8734,
+      "step": 2065408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.617511692827984e-05,
+      "loss": 3.8642,
+      "step": 2065920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.616673098076932e-05,
+      "loss": 3.8692,
+      "step": 2066432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6158361412062533e-05,
+      "loss": 3.8793,
+      "step": 2066944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6149975464552013e-05,
+      "loss": 3.8608,
+      "step": 2067456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6141589517041492e-05,
+      "loss": 3.8629,
+      "step": 2067968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6133203569530972e-05,
+      "loss": 3.8594,
+      "step": 2068480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.612481762202045e-05,
+      "loss": 3.8584,
+      "step": 2068992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.611643167450993e-05,
+      "loss": 3.8656,
+      "step": 2069504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.610806210580314e-05,
+      "loss": 3.8663,
+      "step": 2070016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.609967615829262e-05,
+      "loss": 3.8785,
+      "step": 2070528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.60912902107821e-05,
+      "loss": 3.8775,
+      "step": 2071040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.608290426327158e-05,
+      "loss": 3.869,
+      "step": 2071552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.607451831576106e-05,
+      "loss": 3.8704,
+      "step": 2072064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.606613236825054e-05,
+      "loss": 3.8752,
+      "step": 2072576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.605774642074002e-05,
+      "loss": 3.8543,
+      "step": 2073088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.60493604732295e-05,
+      "loss": 3.8736,
+      "step": 2073600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.604097452571898e-05,
+      "loss": 3.8657,
+      "step": 2074112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.603258857820846e-05,
+      "loss": 3.8539,
+      "step": 2074624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.602420263069794e-05,
+      "loss": 3.8626,
+      "step": 2075136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.601581668318742e-05,
+      "loss": 3.8627,
+      "step": 2075648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.600744711448063e-05,
+      "loss": 3.8657,
+      "step": 2076160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.599906116697011e-05,
+      "loss": 3.8681,
+      "step": 2076672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.599067521945959e-05,
+      "loss": 3.8739,
+      "step": 2077184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5982305650752806e-05,
+      "loss": 3.8704,
+      "step": 2077696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5973936082046015e-05,
+      "loss": 3.8594,
+      "step": 2078208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5965566513339224e-05,
+      "loss": 3.8755,
+      "step": 2078720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5957180565828704e-05,
+      "loss": 3.8762,
+      "step": 2079232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5948794618318187e-05,
+      "loss": 3.8632,
+      "step": 2079744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5940408670807667e-05,
+      "loss": 3.8638,
+      "step": 2080256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5932022723297147e-05,
+      "loss": 3.8628,
+      "step": 2080768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5923653154590356e-05,
+      "loss": 3.8574,
+      "step": 2081280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5915267207079836e-05,
+      "loss": 3.8747,
+      "step": 2081792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5906881259569316e-05,
+      "loss": 3.8722,
+      "step": 2082304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5898495312058796e-05,
+      "loss": 3.8749,
+      "step": 2082816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.589010936454828e-05,
+      "loss": 3.8675,
+      "step": 2083328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5881723417037756e-05,
+      "loss": 3.8688,
+      "step": 2083840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5873337469527236e-05,
+      "loss": 3.8713,
+      "step": 2084352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5864951522016716e-05,
+      "loss": 3.8694,
+      "step": 2084864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5856565574506196e-05,
+      "loss": 3.8437,
+      "step": 2085376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5848179626995676e-05,
+      "loss": 3.878,
+      "step": 2085888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.583981005828889e-05,
+      "loss": 3.8624,
+      "step": 2086400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.583142411077837e-05,
+      "loss": 3.8773,
+      "step": 2086912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.582303816326785e-05,
+      "loss": 3.8663,
+      "step": 2087424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5814652215757328e-05,
+      "loss": 3.8623,
+      "step": 2087936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5806266268246808e-05,
+      "loss": 3.857,
+      "step": 2088448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5797880320736285e-05,
+      "loss": 3.8648,
+      "step": 2088960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5789494373225765e-05,
+      "loss": 3.8791,
+      "step": 2089472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5781108425715245e-05,
+      "loss": 3.8646,
+      "step": 2089984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5772722478204728e-05,
+      "loss": 3.8835,
+      "step": 2090496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5764352909497937e-05,
+      "loss": 3.8686,
+      "step": 2091008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5755966961987417e-05,
+      "loss": 3.8488,
+      "step": 2091520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5747581014476897e-05,
+      "loss": 3.8791,
+      "step": 2092032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5739195066966377e-05,
+      "loss": 3.8594,
+      "step": 2092544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5730809119455857e-05,
+      "loss": 3.8593,
+      "step": 2093056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5722423171945337e-05,
+      "loss": 3.8699,
+      "step": 2093568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5714037224434817e-05,
+      "loss": 3.8717,
+      "step": 2094080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5705651276924297e-05,
+      "loss": 3.8553,
+      "step": 2094592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.569728170821751e-05,
+      "loss": 3.8618,
+      "step": 2095104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.568889576070699e-05,
+      "loss": 3.851,
+      "step": 2095616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5680509813196466e-05,
+      "loss": 3.866,
+      "step": 2096128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5672140244489682e-05,
+      "loss": 3.8706,
+      "step": 2096640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5663754296979162e-05,
+      "loss": 3.8667,
+      "step": 2097152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.565536834946864e-05,
+      "loss": 3.8702,
+      "step": 2097664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5646982401958118e-05,
+      "loss": 3.8609,
+      "step": 2098176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5638596454447598e-05,
+      "loss": 3.8808,
+      "step": 2098688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.563022688574081e-05,
+      "loss": 3.8675,
+      "step": 2099200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.562184093823029e-05,
+      "loss": 3.8771,
+      "step": 2099712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.561345499071977e-05,
+      "loss": 3.8644,
+      "step": 2100224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5605085422012983e-05,
+      "loss": 3.8678,
+      "step": 2100736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5596699474502463e-05,
+      "loss": 3.8712,
+      "step": 2101248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.558831352699194e-05,
+      "loss": 3.8724,
+      "step": 2101760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.557992757948142e-05,
+      "loss": 3.8671,
+      "step": 2102272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5571541631970903e-05,
+      "loss": 3.8771,
+      "step": 2102784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5563155684460383e-05,
+      "loss": 3.8546,
+      "step": 2103296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5554769736949863e-05,
+      "loss": 3.8646,
+      "step": 2103808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5546383789439343e-05,
+      "loss": 3.8636,
+      "step": 2104320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5537997841928823e-05,
+      "loss": 3.875,
+      "step": 2104832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5529628273222032e-05,
+      "loss": 3.8607,
+      "step": 2105344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5521242325711512e-05,
+      "loss": 3.8625,
+      "step": 2105856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5512856378200992e-05,
+      "loss": 3.8661,
+      "step": 2106368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.550447043069047e-05,
+      "loss": 3.8636,
+      "step": 2106880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.549608448317995e-05,
+      "loss": 3.8678,
+      "step": 2107392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.548769853566943e-05,
+      "loss": 3.8577,
+      "step": 2107904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.547931258815891e-05,
+      "loss": 3.8681,
+      "step": 2108416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.547094301945212e-05,
+      "loss": 3.8732,
+      "step": 2108928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5462557071941604e-05,
+      "loss": 3.8725,
+      "step": 2109440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5454171124431084e-05,
+      "loss": 3.847,
+      "step": 2109952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5445801555724293e-05,
+      "loss": 3.8629,
+      "step": 2110464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5437415608213773e-05,
+      "loss": 3.859,
+      "step": 2110976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5429029660703253e-05,
+      "loss": 3.8726,
+      "step": 2111488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5420643713192733e-05,
+      "loss": 3.8729,
+      "step": 2112000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5412257765682213e-05,
+      "loss": 3.8661,
+      "step": 2112512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5403871818171696e-05,
+      "loss": 3.8615,
+      "step": 2113024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5395485870661176e-05,
+      "loss": 3.8645,
+      "step": 2113536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5387099923150653e-05,
+      "loss": 3.8556,
+      "step": 2114048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5378713975640133e-05,
+      "loss": 3.8651,
+      "step": 2114560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5370360785737074e-05,
+      "loss": 3.8629,
+      "step": 2115072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5361974838226558e-05,
+      "loss": 3.8669,
+      "step": 2115584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5353588890716038e-05,
+      "loss": 3.8594,
+      "step": 2116096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5345202943205518e-05,
+      "loss": 3.8519,
+      "step": 2116608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5336816995694998e-05,
+      "loss": 3.8566,
+      "step": 2117120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5328431048184474e-05,
+      "loss": 3.8598,
+      "step": 2117632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5320045100673954e-05,
+      "loss": 3.8663,
+      "step": 2118144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5311659153163434e-05,
+      "loss": 3.8554,
+      "step": 2118656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5303273205652914e-05,
+      "loss": 3.8699,
+      "step": 2119168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5294903636946126e-05,
+      "loss": 3.8667,
+      "step": 2119680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.528653406823934e-05,
+      "loss": 3.8656,
+      "step": 2120192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.527814812072882e-05,
+      "loss": 3.8675,
+      "step": 2120704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.52697621732183e-05,
+      "loss": 3.8608,
+      "step": 2121216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5261376225707775e-05,
+      "loss": 3.8684,
+      "step": 2121728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5252990278197257e-05,
+      "loss": 3.8609,
+      "step": 2122240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5244604330686737e-05,
+      "loss": 3.8596,
+      "step": 2122752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5236218383176219e-05,
+      "loss": 3.8667,
+      "step": 2123264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5227832435665699e-05,
+      "loss": 3.8613,
+      "step": 2123776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5219446488155179e-05,
+      "loss": 3.8678,
+      "step": 2124288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5211060540644657e-05,
+      "loss": 3.8648,
+      "step": 2124800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5202674593134137e-05,
+      "loss": 3.859,
+      "step": 2125312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5194288645623617e-05,
+      "loss": 3.8574,
+      "step": 2125824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.518593545572056e-05,
+      "loss": 3.8543,
+      "step": 2126336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.517754950821004e-05,
+      "loss": 3.8729,
+      "step": 2126848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.516916356069952e-05,
+      "loss": 3.8641,
+      "step": 2127360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5160777613189e-05,
+      "loss": 3.8744,
+      "step": 2127872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5152391665678482e-05,
+      "loss": 3.8584,
+      "step": 2128384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5144005718167958e-05,
+      "loss": 3.8717,
+      "step": 2128896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5135619770657438e-05,
+      "loss": 3.8731,
+      "step": 2129408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.512723382314692e-05,
+      "loss": 3.8619,
+      "step": 2129920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.51188478756364e-05,
+      "loss": 3.8628,
+      "step": 2130432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.511047830692961e-05,
+      "loss": 3.8616,
+      "step": 2130944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5102108738222823e-05,
+      "loss": 3.8737,
+      "step": 2131456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5093722790712303e-05,
+      "loss": 3.8727,
+      "step": 2131968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5085353222005514e-05,
+      "loss": 3.8623,
+      "step": 2132480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5076967274494994e-05,
+      "loss": 3.86,
+      "step": 2132992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5068581326984474e-05,
+      "loss": 3.869,
+      "step": 2133504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5060195379473954e-05,
+      "loss": 3.8724,
+      "step": 2134016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5051809431963432e-05,
+      "loss": 3.857,
+      "step": 2134528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5043423484452912e-05,
+      "loss": 3.8715,
+      "step": 2135040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5035037536942392e-05,
+      "loss": 3.8624,
+      "step": 2135552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5026651589431873e-05,
+      "loss": 3.8716,
+      "step": 2136064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5018265641921353e-05,
+      "loss": 3.8658,
+      "step": 2136576
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.033816337585449,
+      "eval_runtime": 296.7833,
+      "eval_samples_per_second": 1285.756,
+      "eval_steps_per_second": 40.181,
+      "step": 2136960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5009896073214562e-05,
+      "loss": 3.8644,
+      "step": 2137088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5001510125704044e-05,
+      "loss": 3.8598,
+      "step": 2137600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4993124178193524e-05,
+      "loss": 3.8661,
+      "step": 2138112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4984738230683004e-05,
+      "loss": 3.8708,
+      "step": 2138624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4976368661976215e-05,
+      "loss": 3.8736,
+      "step": 2139136
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4967982714465695e-05,
+      "loss": 3.8658,
+      "step": 2139648
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4959596766955175e-05,
+      "loss": 3.8624,
+      "step": 2140160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4951210819444655e-05,
+      "loss": 3.8636,
+      "step": 2140672
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4942841250737865e-05,
+      "loss": 3.8645,
+      "step": 2141184
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4934455303227345e-05,
+      "loss": 3.866,
+      "step": 2141696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4926085734520556e-05,
+      "loss": 3.8649,
+      "step": 2142208
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4917699787010036e-05,
+      "loss": 3.8645,
+      "step": 2142720
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.490933021830325e-05,
+      "loss": 3.8785,
+      "step": 2143232
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4900944270792727e-05,
+      "loss": 3.8548,
+      "step": 2143744
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4892558323282207e-05,
+      "loss": 3.8599,
+      "step": 2144256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4884172375771689e-05,
+      "loss": 3.8561,
+      "step": 2144768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4875786428261168e-05,
+      "loss": 3.8529,
+      "step": 2145280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4867400480750648e-05,
+      "loss": 3.8645,
+      "step": 2145792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.485903091204386e-05,
+      "loss": 3.8588,
+      "step": 2146304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.485064496453334e-05,
+      "loss": 3.8733,
+      "step": 2146816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4842259017022819e-05,
+      "loss": 3.8755,
+      "step": 2147328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4833873069512299e-05,
+      "loss": 3.8641,
+      "step": 2147840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.482548712200178e-05,
+      "loss": 3.867,
+      "step": 2148352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4817101174491257e-05,
+      "loss": 3.8715,
+      "step": 2148864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4808715226980737e-05,
+      "loss": 3.8498,
+      "step": 2149376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4800329279470219e-05,
+      "loss": 3.8713,
+      "step": 2149888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4791943331959699e-05,
+      "loss": 3.8566,
+      "step": 2150400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4783557384449179e-05,
+      "loss": 3.8538,
+      "step": 2150912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4775171436938659e-05,
+      "loss": 3.8612,
+      "step": 2151424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4766785489428139e-05,
+      "loss": 3.855,
+      "step": 2151936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4758399541917617e-05,
+      "loss": 3.8622,
+      "step": 2152448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.475002997321083e-05,
+      "loss": 3.8614,
+      "step": 2152960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.474164402570031e-05,
+      "loss": 3.872,
+      "step": 2153472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4733258078189788e-05,
+      "loss": 3.8642,
+      "step": 2153984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4724872130679268e-05,
+      "loss": 3.8592,
+      "step": 2154496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4716502561972482e-05,
+      "loss": 3.8677,
+      "step": 2155008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4708116614461962e-05,
+      "loss": 3.8715,
+      "step": 2155520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4699730666951438e-05,
+      "loss": 3.859,
+      "step": 2156032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.469134471944092e-05,
+      "loss": 3.8559,
+      "step": 2156544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.46829587719304e-05,
+      "loss": 3.8598,
+      "step": 2157056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.467457282441988e-05,
+      "loss": 3.8563,
+      "step": 2157568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.466618687690936e-05,
+      "loss": 3.8669,
+      "step": 2158080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.465780092939884e-05,
+      "loss": 3.8652,
+      "step": 2158592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4649414981888321e-05,
+      "loss": 3.8734,
+      "step": 2159104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.464104541318153e-05,
+      "loss": 3.8676,
+      "step": 2159616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4632659465671012e-05,
+      "loss": 3.8611,
+      "step": 2160128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4624273518160492e-05,
+      "loss": 3.8663,
+      "step": 2160640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4615887570649969e-05,
+      "loss": 3.8655,
+      "step": 2161152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.460750162313945e-05,
+      "loss": 3.8413,
+      "step": 2161664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4599132054432663e-05,
+      "loss": 3.8713,
+      "step": 2162176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4590746106922143e-05,
+      "loss": 3.8614,
+      "step": 2162688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4582360159411621e-05,
+      "loss": 3.8726,
+      "step": 2163200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4573974211901101e-05,
+      "loss": 3.8595,
+      "step": 2163712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4565588264390581e-05,
+      "loss": 3.8638,
+      "step": 2164224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4557202316880061e-05,
+      "loss": 3.849,
+      "step": 2164736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4548816369369541e-05,
+      "loss": 3.8589,
+      "step": 2165248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4540430421859023e-05,
+      "loss": 3.8765,
+      "step": 2165760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4532060853152232e-05,
+      "loss": 3.8611,
+      "step": 2166272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4523674905641713e-05,
+      "loss": 3.8768,
+      "step": 2166784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4515288958131193e-05,
+      "loss": 3.8659,
+      "step": 2167296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4506903010620673e-05,
+      "loss": 3.8408,
+      "step": 2167808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4498533441913884e-05,
+      "loss": 3.8785,
+      "step": 2168320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4490147494403364e-05,
+      "loss": 3.8521,
+      "step": 2168832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4481761546892844e-05,
+      "loss": 3.8563,
+      "step": 2169344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4473375599382324e-05,
+      "loss": 3.867,
+      "step": 2169856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4465006030675535e-05,
+      "loss": 3.8675,
+      "step": 2170368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4456636461968745e-05,
+      "loss": 3.8547,
+      "step": 2170880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4448250514458225e-05,
+      "loss": 3.8559,
+      "step": 2171392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4439864566947705e-05,
+      "loss": 3.8508,
+      "step": 2171904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4431478619437185e-05,
+      "loss": 3.8559,
+      "step": 2172416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4423092671926667e-05,
+      "loss": 3.871,
+      "step": 2172928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4414739482023609e-05,
+      "loss": 3.8651,
+      "step": 2173440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.440635353451309e-05,
+      "loss": 3.8628,
+      "step": 2173952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4397967587002567e-05,
+      "loss": 3.8567,
+      "step": 2174464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4389581639492047e-05,
+      "loss": 3.8775,
+      "step": 2174976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4381195691981528e-05,
+      "loss": 3.8624,
+      "step": 2175488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4372809744471008e-05,
+      "loss": 3.8707,
+      "step": 2176000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4364423796960488e-05,
+      "loss": 3.858,
+      "step": 2176512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4356037849449968e-05,
+      "loss": 3.8646,
+      "step": 2177024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4347651901939448e-05,
+      "loss": 3.8689,
+      "step": 2177536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4339265954428927e-05,
+      "loss": 3.8659,
+      "step": 2178048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4330880006918406e-05,
+      "loss": 3.8638,
+      "step": 2178560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4322494059407886e-05,
+      "loss": 3.8711,
+      "step": 2179072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4314124490701097e-05,
+      "loss": 3.8525,
+      "step": 2179584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4305738543190577e-05,
+      "loss": 3.8558,
+      "step": 2180096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4297352595680059e-05,
+      "loss": 3.8602,
+      "step": 2180608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4288983026973271e-05,
+      "loss": 3.8688,
+      "step": 2181120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.4280597079462748e-05,
+      "loss": 3.8581,
+      "step": 2181632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.427221113195223e-05,
+      "loss": 3.8638,
+      "step": 2182144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.426382518444171e-05,
+      "loss": 3.8603,
+      "step": 2182656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.425545561573492e-05,
+      "loss": 3.8592,
+      "step": 2183168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.42470696682244e-05,
+      "loss": 3.8623,
+      "step": 2183680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.423868372071388e-05,
+      "loss": 3.8541,
+      "step": 2184192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4230314152007093e-05,
+      "loss": 3.8642,
+      "step": 2184704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4221928204496571e-05,
+      "loss": 3.869,
+      "step": 2185216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4213542256986051e-05,
+      "loss": 3.8725,
+      "step": 2185728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.420515630947553e-05,
+      "loss": 3.8372,
+      "step": 2186240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4196770361965012e-05,
+      "loss": 3.8589,
+      "step": 2186752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4188384414454492e-05,
+      "loss": 3.8536,
+      "step": 2187264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4179998466943972e-05,
+      "loss": 3.8677,
+      "step": 2187776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4171612519433452e-05,
+      "loss": 3.8691,
+      "step": 2188288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4163242950726663e-05,
+      "loss": 3.8602,
+      "step": 2188800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4154857003216143e-05,
+      "loss": 3.8558,
+      "step": 2189312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4146471055705623e-05,
+      "loss": 3.8609,
+      "step": 2189824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4138085108195101e-05,
+      "loss": 3.8567,
+      "step": 2190336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4129699160684581e-05,
+      "loss": 3.8607,
+      "step": 2190848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4121329591977794e-05,
+      "loss": 3.8552,
+      "step": 2191360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4112943644467275e-05,
+      "loss": 3.8634,
+      "step": 2191872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4104557696956752e-05,
+      "loss": 3.8553,
+      "step": 2192384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4096171749446232e-05,
+      "loss": 3.8511,
+      "step": 2192896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4087785801935714e-05,
+      "loss": 3.853,
+      "step": 2193408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4079416233228926e-05,
+      "loss": 3.8495,
+      "step": 2193920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4071030285718403e-05,
+      "loss": 3.8634,
+      "step": 2194432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4062644338207884e-05,
+      "loss": 3.8501,
+      "step": 2194944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4054258390697364e-05,
+      "loss": 3.8674,
+      "step": 2195456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4045872443186844e-05,
+      "loss": 3.8631,
+      "step": 2195968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4037486495676324e-05,
+      "loss": 3.8615,
+      "step": 2196480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4029116926969535e-05,
+      "loss": 3.8605,
+      "step": 2196992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4020730979459015e-05,
+      "loss": 3.8591,
+      "step": 2197504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4012345031948495e-05,
+      "loss": 3.8664,
+      "step": 2198016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4003959084437976e-05,
+      "loss": 3.8565,
+      "step": 2198528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3995573136927456e-05,
+      "loss": 3.8539,
+      "step": 2199040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3987187189416933e-05,
+      "loss": 3.8611,
+      "step": 2199552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3978817620710147e-05,
+      "loss": 3.8586,
+      "step": 2200064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3970431673199627e-05,
+      "loss": 3.8622,
+      "step": 2200576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3962045725689107e-05,
+      "loss": 3.8558,
+      "step": 2201088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3953659778178585e-05,
+      "loss": 3.8561,
+      "step": 2201600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3945273830668065e-05,
+      "loss": 3.8581,
+      "step": 2202112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3936904261961278e-05,
+      "loss": 3.8524,
+      "step": 2202624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3928518314450756e-05,
+      "loss": 3.8623,
+      "step": 2203136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3920132366940236e-05,
+      "loss": 3.8614,
+      "step": 2203648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3911746419429716e-05,
+      "loss": 3.8729,
+      "step": 2204160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3903360471919196e-05,
+      "loss": 3.8555,
+      "step": 2204672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3894974524408678e-05,
+      "loss": 3.8629,
+      "step": 2205184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3886588576898158e-05,
+      "loss": 3.8682,
+      "step": 2205696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3878202629387637e-05,
+      "loss": 3.8596,
+      "step": 2206208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3869833060680848e-05,
+      "loss": 3.8561,
+      "step": 2206720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3861447113170328e-05,
+      "loss": 3.8587,
+      "step": 2207232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3853061165659808e-05,
+      "loss": 3.8724,
+      "step": 2207744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3844675218149288e-05,
+      "loss": 3.8678,
+      "step": 2208256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3836305649442499e-05,
+      "loss": 3.8601,
+      "step": 2208768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3827919701931979e-05,
+      "loss": 3.8541,
+      "step": 2209280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.381953375442146e-05,
+      "loss": 3.8641,
+      "step": 2209792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3811147806910937e-05,
+      "loss": 3.8684,
+      "step": 2210304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.380277823820415e-05,
+      "loss": 3.8497,
+      "step": 2210816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3794392290693631e-05,
+      "loss": 3.8732,
+      "step": 2211328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3786006343183111e-05,
+      "loss": 3.8535,
+      "step": 2211840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3777620395672588e-05,
+      "loss": 3.8662,
+      "step": 2212352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3769250826965802e-05,
+      "loss": 3.8636,
+      "step": 2212864
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.032118797302246,
+      "eval_runtime": 294.2323,
+      "eval_samples_per_second": 1296.904,
+      "eval_steps_per_second": 40.529,
+      "step": 2213280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3760864879455282e-05,
+      "loss": 3.884,
+      "step": 2213376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3752478931944762e-05,
+      "loss": 3.8561,
+      "step": 2213888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.374409298443424e-05,
+      "loss": 3.8605,
+      "step": 2214400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.373570703692372e-05,
+      "loss": 3.8703,
+      "step": 2214912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.37273210894132e-05,
+      "loss": 3.8623,
+      "step": 2215424
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.371893514190268e-05,
+      "loss": 3.8667,
+      "step": 2215936
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3710549194392162e-05,
+      "loss": 3.8573,
+      "step": 2216448
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3702163246881642e-05,
+      "loss": 3.855,
+      "step": 2216960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3693777299371118e-05,
+      "loss": 3.8594,
+      "step": 2217472
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.36853913518606e-05,
+      "loss": 3.8658,
+      "step": 2217984
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.367700540435008e-05,
+      "loss": 3.8581,
+      "step": 2218496
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.366861945683956e-05,
+      "loss": 3.8668,
+      "step": 2219008
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.366024988813277e-05,
+      "loss": 3.8705,
+      "step": 2219520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.365186394062225e-05,
+      "loss": 3.8492,
+      "step": 2220032
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.364347799311173e-05,
+      "loss": 3.8602,
+      "step": 2220544
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.363509204560121e-05,
+      "loss": 3.8475,
+      "step": 2221056
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3626706098090692e-05,
+      "loss": 3.8478,
+      "step": 2221568
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3618320150580172e-05,
+      "loss": 3.8635,
+      "step": 2222080
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3609934203069649e-05,
+      "loss": 3.8519,
+      "step": 2222592
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.360154825555913e-05,
+      "loss": 3.8683,
+      "step": 2223104
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.359316230804861e-05,
+      "loss": 3.8712,
+      "step": 2223616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.358477636053809e-05,
+      "loss": 3.8601,
+      "step": 2224128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.357639041302757e-05,
+      "loss": 3.8659,
+      "step": 2224640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3568020844320781e-05,
+      "loss": 3.8681,
+      "step": 2225152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.355963489681026e-05,
+      "loss": 3.8483,
+      "step": 2225664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.355124894929974e-05,
+      "loss": 3.8637,
+      "step": 2226176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3542863001789222e-05,
+      "loss": 3.8558,
+      "step": 2226688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3534477054278702e-05,
+      "loss": 3.8523,
+      "step": 2227200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3526091106768182e-05,
+      "loss": 3.8591,
+      "step": 2227712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.351770515925766e-05,
+      "loss": 3.8496,
+      "step": 2228224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.350931921174714e-05,
+      "loss": 3.8554,
+      "step": 2228736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3500949643040353e-05,
+      "loss": 3.8584,
+      "step": 2229248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3492563695529831e-05,
+      "loss": 3.8671,
+      "step": 2229760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3484194126823044e-05,
+      "loss": 3.8587,
+      "step": 2230272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3475808179312524e-05,
+      "loss": 3.8556,
+      "step": 2230784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3467438610605735e-05,
+      "loss": 3.8587,
+      "step": 2231296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3459052663095214e-05,
+      "loss": 3.8706,
+      "step": 2231808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3450666715584694e-05,
+      "loss": 3.8565,
+      "step": 2232320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3442280768074176e-05,
+      "loss": 3.8545,
+      "step": 2232832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3433894820563656e-05,
+      "loss": 3.855,
+      "step": 2233344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3425508873053133e-05,
+      "loss": 3.8576,
+      "step": 2233856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3417122925542613e-05,
+      "loss": 3.858,
+      "step": 2234368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3408736978032094e-05,
+      "loss": 3.8658,
+      "step": 2234880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3400351030521574e-05,
+      "loss": 3.8657,
+      "step": 2235392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3391965083011054e-05,
+      "loss": 3.8614,
+      "step": 2235904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3383579135500534e-05,
+      "loss": 3.8592,
+      "step": 2236416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3375193187990012e-05,
+      "loss": 3.8599,
+      "step": 2236928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3366839998086956e-05,
+      "loss": 3.8615,
+      "step": 2237440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3358454050576436e-05,
+      "loss": 3.8414,
+      "step": 2237952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3350068103065916e-05,
+      "loss": 3.8654,
+      "step": 2238464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3341682155555396e-05,
+      "loss": 3.8574,
+      "step": 2238976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3333296208044877e-05,
+      "loss": 3.867,
+      "step": 2239488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3324910260534357e-05,
+      "loss": 3.8554,
+      "step": 2240000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3316524313023837e-05,
+      "loss": 3.8569,
+      "step": 2240512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3308154744317048e-05,
+      "loss": 3.8514,
+      "step": 2241024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3299768796806528e-05,
+      "loss": 3.8487,
+      "step": 2241536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3291382849296008e-05,
+      "loss": 3.8719,
+      "step": 2242048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3282996901785486e-05,
+      "loss": 3.8557,
+      "step": 2242560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3274610954274966e-05,
+      "loss": 3.8779,
+      "step": 2243072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3266241385568179e-05,
+      "loss": 3.8569,
+      "step": 2243584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3257855438057658e-05,
+      "loss": 3.8415,
+      "step": 2244096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3249469490547137e-05,
+      "loss": 3.875,
+      "step": 2244608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3241083543036617e-05,
+      "loss": 3.8518,
+      "step": 2245120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3232697595526097e-05,
+      "loss": 3.8507,
+      "step": 2245632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.322432802681931e-05,
+      "loss": 3.8567,
+      "step": 2246144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3215942079308787e-05,
+      "loss": 3.8674,
+      "step": 2246656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3207556131798269e-05,
+      "loss": 3.848,
+      "step": 2247168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3199170184287749e-05,
+      "loss": 3.8544,
+      "step": 2247680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3190784236777229e-05,
+      "loss": 3.8468,
+      "step": 2248192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.318241466807044e-05,
+      "loss": 3.8473,
+      "step": 2248704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.317402872055992e-05,
+      "loss": 3.8697,
+      "step": 2249216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.31656427730494e-05,
+      "loss": 3.8626,
+      "step": 2249728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.315725682553888e-05,
+      "loss": 3.8592,
+      "step": 2250240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.314887087802836e-05,
+      "loss": 3.8537,
+      "step": 2250752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3140484930517841e-05,
+      "loss": 3.8716,
+      "step": 2251264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3132098983007318e-05,
+      "loss": 3.859,
+      "step": 2251776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3123729414300532e-05,
+      "loss": 3.868,
+      "step": 2252288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3115343466790012e-05,
+      "loss": 3.8554,
+      "step": 2252800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3106957519279488e-05,
+      "loss": 3.8583,
+      "step": 2253312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3098587950572703e-05,
+      "loss": 3.8657,
+      "step": 2253824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3090202003062183e-05,
+      "loss": 3.8572,
+      "step": 2254336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3081816055551663e-05,
+      "loss": 3.8612,
+      "step": 2254848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.307343010804114e-05,
+      "loss": 3.8683,
+      "step": 2255360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.306504416053062e-05,
+      "loss": 3.8551,
+      "step": 2255872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.30566582130201e-05,
+      "loss": 3.8496,
+      "step": 2256384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.304827226550958e-05,
+      "loss": 3.8573,
+      "step": 2256896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3039886317999062e-05,
+      "loss": 3.8662,
+      "step": 2257408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3031516749292271e-05,
+      "loss": 3.8568,
+      "step": 2257920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3023130801781751e-05,
+      "loss": 3.8552,
+      "step": 2258432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3014744854271233e-05,
+      "loss": 3.8542,
+      "step": 2258944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3006358906760713e-05,
+      "loss": 3.8549,
+      "step": 2259456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2997972959250193e-05,
+      "loss": 3.859,
+      "step": 2259968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2989587011739671e-05,
+      "loss": 3.8548,
+      "step": 2260480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2981201064229151e-05,
+      "loss": 3.859,
+      "step": 2260992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2972815116718631e-05,
+      "loss": 3.862,
+      "step": 2261504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2964445548011844e-05,
+      "loss": 3.8689,
+      "step": 2262016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2956059600501322e-05,
+      "loss": 3.8384,
+      "step": 2262528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2947690031794534e-05,
+      "loss": 3.853,
+      "step": 2263040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2939304084284016e-05,
+      "loss": 3.8526,
+      "step": 2263552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2930918136773496e-05,
+      "loss": 3.8627,
+      "step": 2264064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2922532189262973e-05,
+      "loss": 3.8671,
+      "step": 2264576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2914162620556187e-05,
+      "loss": 3.8552,
+      "step": 2265088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2905793051849396e-05,
+      "loss": 3.8519,
+      "step": 2265600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2897407104338877e-05,
+      "loss": 3.8583,
+      "step": 2266112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2889021156828357e-05,
+      "loss": 3.8546,
+      "step": 2266624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2880635209317837e-05,
+      "loss": 3.8547,
+      "step": 2267136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2872249261807317e-05,
+      "loss": 3.8516,
+      "step": 2267648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2863863314296796e-05,
+      "loss": 3.8602,
+      "step": 2268160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2855477366786276e-05,
+      "loss": 3.8496,
+      "step": 2268672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2847091419275755e-05,
+      "loss": 3.8479,
+      "step": 2269184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2838721850568968e-05,
+      "loss": 3.8508,
+      "step": 2269696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2830335903058446e-05,
+      "loss": 3.8424,
+      "step": 2270208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2821949955547926e-05,
+      "loss": 3.8601,
+      "step": 2270720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2813564008037406e-05,
+      "loss": 3.8517,
+      "step": 2271232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2805178060526888e-05,
+      "loss": 3.855,
+      "step": 2271744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2796808491820097e-05,
+      "loss": 3.8613,
+      "step": 2272256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2788422544309579e-05,
+      "loss": 3.8595,
+      "step": 2272768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2780036596799058e-05,
+      "loss": 3.8555,
+      "step": 2273280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2771650649288538e-05,
+      "loss": 3.8529,
+      "step": 2273792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2763264701778018e-05,
+      "loss": 3.8626,
+      "step": 2274304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2754878754267498e-05,
+      "loss": 3.8517,
+      "step": 2274816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2746492806756977e-05,
+      "loss": 3.8476,
+      "step": 2275328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2738106859246457e-05,
+      "loss": 3.8616,
+      "step": 2275840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.27297536693434e-05,
+      "loss": 3.8542,
+      "step": 2276352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.272136772183288e-05,
+      "loss": 3.8579,
+      "step": 2276864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.271298177432236e-05,
+      "loss": 3.8523,
+      "step": 2277376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2704595826811841e-05,
+      "loss": 3.8553,
+      "step": 2277888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2696209879301321e-05,
+      "loss": 3.8543,
+      "step": 2278400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2687840310594532e-05,
+      "loss": 3.8499,
+      "step": 2278912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2679454363084012e-05,
+      "loss": 3.851,
+      "step": 2279424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2671068415573492e-05,
+      "loss": 3.8641,
+      "step": 2279936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2662682468062972e-05,
+      "loss": 3.868,
+      "step": 2280448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2654312899356183e-05,
+      "loss": 3.8504,
+      "step": 2280960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2645926951845663e-05,
+      "loss": 3.8535,
+      "step": 2281472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2637557383138874e-05,
+      "loss": 3.8695,
+      "step": 2281984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2629171435628354e-05,
+      "loss": 3.8564,
+      "step": 2282496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2620785488117834e-05,
+      "loss": 3.8502,
+      "step": 2283008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2612399540607313e-05,
+      "loss": 3.8534,
+      "step": 2283520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2604013593096795e-05,
+      "loss": 3.8689,
+      "step": 2284032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2595627645586272e-05,
+      "loss": 3.8652,
+      "step": 2284544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2587241698075752e-05,
+      "loss": 3.8553,
+      "step": 2285056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2578855750565233e-05,
+      "loss": 3.8479,
+      "step": 2285568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2570469803054713e-05,
+      "loss": 3.8632,
+      "step": 2286080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2562083855544193e-05,
+      "loss": 3.8623,
+      "step": 2286592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2553714286837404e-05,
+      "loss": 3.8479,
+      "step": 2287104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2545328339326884e-05,
+      "loss": 3.8681,
+      "step": 2287616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2536942391816364e-05,
+      "loss": 3.8528,
+      "step": 2288128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2528556444305844e-05,
+      "loss": 3.8621,
+      "step": 2288640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2520170496795326e-05,
+      "loss": 3.8603,
+      "step": 2289152
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.031382083892822,
+      "eval_runtime": 279.3155,
+      "eval_samples_per_second": 1366.165,
+      "eval_steps_per_second": 42.694,
+      "step": 2289600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2511784549284802e-05,
+      "loss": 3.858,
+      "step": 2289664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2503414980578015e-05,
+      "loss": 3.8491,
+      "step": 2290176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2495029033067495e-05,
+      "loss": 3.8574,
+      "step": 2290688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2486643085556974e-05,
+      "loss": 3.8664,
+      "step": 2291200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2478257138046454e-05,
+      "loss": 3.8603,
+      "step": 2291712
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2469871190535936e-05,
+      "loss": 3.8584,
+      "step": 2292224
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2461485243025414e-05,
+      "loss": 3.8561,
+      "step": 2292736
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2453099295514894e-05,
+      "loss": 3.8533,
+      "step": 2293248
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2444713348004374e-05,
+      "loss": 3.8539,
+      "step": 2293760
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2436343779297585e-05,
+      "loss": 3.8617,
+      "step": 2294272
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2427957831787065e-05,
+      "loss": 3.8543,
+      "step": 2294784
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2419588263080277e-05,
+      "loss": 3.8658,
+      "step": 2295296
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2411218694373488e-05,
+      "loss": 3.8662,
+      "step": 2295808
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2402832746862968e-05,
+      "loss": 3.846,
+      "step": 2296320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2394446799352448e-05,
+      "loss": 3.8542,
+      "step": 2296832
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2386060851841928e-05,
+      "loss": 3.8427,
+      "step": 2297344
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2377674904331408e-05,
+      "loss": 3.8467,
+      "step": 2297856
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2369288956820888e-05,
+      "loss": 3.8578,
+      "step": 2298368
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2360903009310368e-05,
+      "loss": 3.8473,
+      "step": 2298880
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2352517061799848e-05,
+      "loss": 3.8613,
+      "step": 2299392
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2344131114289326e-05,
+      "loss": 3.8699,
+      "step": 2299904
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2335745166778808e-05,
+      "loss": 3.8523,
+      "step": 2300416
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2327359219268288e-05,
+      "loss": 3.862,
+      "step": 2300928
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2318989650561499e-05,
+      "loss": 3.8625,
+      "step": 2301440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2310603703050979e-05,
+      "loss": 3.8457,
+      "step": 2301952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2302217755540459e-05,
+      "loss": 3.8619,
+      "step": 2302464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2293831808029938e-05,
+      "loss": 3.8488,
+      "step": 2302976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2285445860519418e-05,
+      "loss": 3.8526,
+      "step": 2303488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2277059913008898e-05,
+      "loss": 3.854,
+      "step": 2304000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.226869034430211e-05,
+      "loss": 3.8441,
+      "step": 2304512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2260304396791591e-05,
+      "loss": 3.8509,
+      "step": 2305024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2251918449281069e-05,
+      "loss": 3.8589,
+      "step": 2305536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2243532501770549e-05,
+      "loss": 3.8641,
+      "step": 2306048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2235162933063762e-05,
+      "loss": 3.8556,
+      "step": 2306560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.222677698555324e-05,
+      "loss": 3.8491,
+      "step": 2307072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2218407416846452e-05,
+      "loss": 3.8546,
+      "step": 2307584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2210021469335932e-05,
+      "loss": 3.8707,
+      "step": 2308096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2201635521825412e-05,
+      "loss": 3.8483,
+      "step": 2308608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.219324957431489e-05,
+      "loss": 3.8524,
+      "step": 2309120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2184863626804372e-05,
+      "loss": 3.8533,
+      "step": 2309632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2176477679293852e-05,
+      "loss": 3.8507,
+      "step": 2310144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.216809173178333e-05,
+      "loss": 3.8572,
+      "step": 2310656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.215970578427281e-05,
+      "loss": 3.8597,
+      "step": 2311168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2151319836762292e-05,
+      "loss": 3.8649,
+      "step": 2311680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2142933889251772e-05,
+      "loss": 3.8573,
+      "step": 2312192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2134564320544983e-05,
+      "loss": 3.8541,
+      "step": 2312704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2126178373034463e-05,
+      "loss": 3.8523,
+      "step": 2313216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2117792425523943e-05,
+      "loss": 3.863,
+      "step": 2313728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2109406478013421e-05,
+      "loss": 3.8356,
+      "step": 2314240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2101036909306633e-05,
+      "loss": 3.8578,
+      "step": 2314752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2092650961796113e-05,
+      "loss": 3.8598,
+      "step": 2315264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2084265014285593e-05,
+      "loss": 3.8619,
+      "step": 2315776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2075879066775073e-05,
+      "loss": 3.8532,
+      "step": 2316288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2067493119264553e-05,
+      "loss": 3.8472,
+      "step": 2316800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2059107171754033e-05,
+      "loss": 3.8537,
+      "step": 2317312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2050737603047246e-05,
+      "loss": 3.8456,
+      "step": 2317824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2042351655536724e-05,
+      "loss": 3.8671,
+      "step": 2318336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2033965708026204e-05,
+      "loss": 3.8538,
+      "step": 2318848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2025579760515684e-05,
+      "loss": 3.8719,
+      "step": 2319360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2017193813005164e-05,
+      "loss": 3.853,
+      "step": 2319872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2008807865494644e-05,
+      "loss": 3.8388,
+      "step": 2320384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2000421917984124e-05,
+      "loss": 3.8697,
+      "step": 2320896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1992035970473604e-05,
+      "loss": 3.8468,
+      "step": 2321408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1983650022963084e-05,
+      "loss": 3.8471,
+      "step": 2321920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1975264075452564e-05,
+      "loss": 3.8561,
+      "step": 2322432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1966894506745776e-05,
+      "loss": 3.8618,
+      "step": 2322944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1958508559235254e-05,
+      "loss": 3.8426,
+      "step": 2323456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1950122611724734e-05,
+      "loss": 3.8506,
+      "step": 2323968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1941736664214214e-05,
+      "loss": 3.8471,
+      "step": 2324480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1933350716703694e-05,
+      "loss": 3.8434,
+      "step": 2324992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1924964769193174e-05,
+      "loss": 3.864,
+      "step": 2325504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1916595200486385e-05,
+      "loss": 3.8606,
+      "step": 2326016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1908209252975867e-05,
+      "loss": 3.853,
+      "step": 2326528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1899823305465345e-05,
+      "loss": 3.8511,
+      "step": 2327040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1891437357954825e-05,
+      "loss": 3.865,
+      "step": 2327552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1883051410444305e-05,
+      "loss": 3.854,
+      "step": 2328064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1874681841737515e-05,
+      "loss": 3.861,
+      "step": 2328576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1866312273030728e-05,
+      "loss": 3.856,
+      "step": 2329088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1857942704323939e-05,
+      "loss": 3.8552,
+      "step": 2329600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1849556756813419e-05,
+      "loss": 3.8602,
+      "step": 2330112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1841170809302899e-05,
+      "loss": 3.8574,
+      "step": 2330624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1832784861792379e-05,
+      "loss": 3.8573,
+      "step": 2331136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1824398914281859e-05,
+      "loss": 3.8625,
+      "step": 2331648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1816012966771339e-05,
+      "loss": 3.8581,
+      "step": 2332160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1807627019260818e-05,
+      "loss": 3.8421,
+      "step": 2332672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1799241071750298e-05,
+      "loss": 3.8537,
+      "step": 2333184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1790855124239778e-05,
+      "loss": 3.8612,
+      "step": 2333696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1782469176729258e-05,
+      "loss": 3.8525,
+      "step": 2334208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1774083229218738e-05,
+      "loss": 3.854,
+      "step": 2334720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1765713660511949e-05,
+      "loss": 3.851,
+      "step": 2335232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.175732771300143e-05,
+      "loss": 3.8518,
+      "step": 2335744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1748941765490909e-05,
+      "loss": 3.8515,
+      "step": 2336256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1740555817980389e-05,
+      "loss": 3.8519,
+      "step": 2336768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1732169870469869e-05,
+      "loss": 3.857,
+      "step": 2337280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1723783922959349e-05,
+      "loss": 3.8554,
+      "step": 2337792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1715397975448829e-05,
+      "loss": 3.8671,
+      "step": 2338304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1707012027938309e-05,
+      "loss": 3.8301,
+      "step": 2338816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1698626080427789e-05,
+      "loss": 3.8535,
+      "step": 2339328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1690240132917269e-05,
+      "loss": 3.8488,
+      "step": 2339840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1681854185406749e-05,
+      "loss": 3.8593,
+      "step": 2340352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1673468237896229e-05,
+      "loss": 3.8622,
+      "step": 2340864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.166511504799317e-05,
+      "loss": 3.8539,
+      "step": 2341376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.165672910048265e-05,
+      "loss": 3.8475,
+      "step": 2341888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1648343152972132e-05,
+      "loss": 3.8545,
+      "step": 2342400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1639957205461612e-05,
+      "loss": 3.8553,
+      "step": 2342912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1631587636754823e-05,
+      "loss": 3.8469,
+      "step": 2343424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1623201689244303e-05,
+      "loss": 3.8467,
+      "step": 2343936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1614815741733783e-05,
+      "loss": 3.8523,
+      "step": 2344448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.160642979422326e-05,
+      "loss": 3.8536,
+      "step": 2344960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1598043846712742e-05,
+      "loss": 3.8418,
+      "step": 2345472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1589674278005953e-05,
+      "loss": 3.8519,
+      "step": 2345984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1581288330495433e-05,
+      "loss": 3.8394,
+      "step": 2346496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1572902382984913e-05,
+      "loss": 3.8583,
+      "step": 2347008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1564516435474393e-05,
+      "loss": 3.8451,
+      "step": 2347520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1556146866767604e-05,
+      "loss": 3.8526,
+      "step": 2348032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1547760919257086e-05,
+      "loss": 3.8548,
+      "step": 2348544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1539374971746564e-05,
+      "loss": 3.8545,
+      "step": 2349056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1530989024236044e-05,
+      "loss": 3.853,
+      "step": 2349568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1522603076725524e-05,
+      "loss": 3.8501,
+      "step": 2350080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1514233508018734e-05,
+      "loss": 3.8583,
+      "step": 2350592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1505863939311947e-05,
+      "loss": 3.8462,
+      "step": 2351104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1497477991801427e-05,
+      "loss": 3.8433,
+      "step": 2351616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1489092044290907e-05,
+      "loss": 3.859,
+      "step": 2352128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1480706096780385e-05,
+      "loss": 3.8556,
+      "step": 2352640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1472320149269867e-05,
+      "loss": 3.8545,
+      "step": 2353152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1463934201759347e-05,
+      "loss": 3.8467,
+      "step": 2353664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1455548254248825e-05,
+      "loss": 3.8522,
+      "step": 2354176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1447162306738305e-05,
+      "loss": 3.8479,
+      "step": 2354688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1438776359227787e-05,
+      "loss": 3.8459,
+      "step": 2355200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1430390411717267e-05,
+      "loss": 3.8433,
+      "step": 2355712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1422004464206745e-05,
+      "loss": 3.8626,
+      "step": 2356224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1413618516696225e-05,
+      "loss": 3.8654,
+      "step": 2356736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1405248947989437e-05,
+      "loss": 3.849,
+      "step": 2357248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1396863000478916e-05,
+      "loss": 3.8503,
+      "step": 2357760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1388477052968397e-05,
+      "loss": 3.86,
+      "step": 2358272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1380091105457877e-05,
+      "loss": 3.8531,
+      "step": 2358784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1371721536751088e-05,
+      "loss": 3.8488,
+      "step": 2359296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1363351968044299e-05,
+      "loss": 3.854,
+      "step": 2359808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1354966020533779e-05,
+      "loss": 3.8635,
+      "step": 2360320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1346580073023259e-05,
+      "loss": 3.8599,
+      "step": 2360832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.133819412551274e-05,
+      "loss": 3.8537,
+      "step": 2361344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1329808178002219e-05,
+      "loss": 3.8448,
+      "step": 2361856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1321422230491698e-05,
+      "loss": 3.8594,
+      "step": 2362368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1313036282981178e-05,
+      "loss": 3.8593,
+      "step": 2362880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.130466671427439e-05,
+      "loss": 3.8447,
+      "step": 2363392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.129628076676387e-05,
+      "loss": 3.8676,
+      "step": 2363904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.128789481925335e-05,
+      "loss": 3.8496,
+      "step": 2364416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1279508871742829e-05,
+      "loss": 3.8613,
+      "step": 2364928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1271122924232309e-05,
+      "loss": 3.8524,
+      "step": 2365440
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.030492305755615,
+      "eval_runtime": 279.5706,
+      "eval_samples_per_second": 1364.918,
+      "eval_steps_per_second": 42.655,
+      "step": 2365920
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1262753355525522e-05,
+      "loss": 3.8525,
+      "step": 2365952
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1254367408015002e-05,
+      "loss": 3.8496,
+      "step": 2366464
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.124598146050448e-05,
+      "loss": 3.8536,
+      "step": 2366976
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1237595512993961e-05,
+      "loss": 3.8608,
+      "step": 2367488
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1229209565483441e-05,
+      "loss": 3.8563,
+      "step": 2368000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.122082361797292e-05,
+      "loss": 3.857,
+      "step": 2368512
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.12124376704624e-05,
+      "loss": 3.852,
+      "step": 2369024
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1204051722951881e-05,
+      "loss": 3.8518,
+      "step": 2369536
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1195682154245092e-05,
+      "loss": 3.8524,
+      "step": 2370048
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.118729620673457e-05,
+      "loss": 3.8603,
+      "step": 2370560
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1178910259224052e-05,
+      "loss": 3.8496,
+      "step": 2371072
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1170524311713532e-05,
+      "loss": 3.8575,
+      "step": 2371584
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1162171121810475e-05,
+      "loss": 3.8643,
+      "step": 2372096
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1153785174299953e-05,
+      "loss": 3.8439,
+      "step": 2372608
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1145399226789433e-05,
+      "loss": 3.8521,
+      "step": 2373120
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1137013279278913e-05,
+      "loss": 3.837,
+      "step": 2373632
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1128627331768393e-05,
+      "loss": 3.8451,
+      "step": 2374144
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1120257763061604e-05,
+      "loss": 3.8526,
+      "step": 2374656
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1111871815551086e-05,
+      "loss": 3.8447,
+      "step": 2375168
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1103485868040566e-05,
+      "loss": 3.8539,
+      "step": 2375680
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1095099920530044e-05,
+      "loss": 3.8691,
+      "step": 2376192
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1086713973019524e-05,
+      "loss": 3.8502,
+      "step": 2376704
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1078328025509006e-05,
+      "loss": 3.8572,
+      "step": 2377216
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1069958456802216e-05,
+      "loss": 3.8612,
+      "step": 2377728
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1061572509291696e-05,
+      "loss": 3.848,
+      "step": 2378240
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1053186561781176e-05,
+      "loss": 3.8531,
+      "step": 2378752
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1044800614270656e-05,
+      "loss": 3.8397,
+      "step": 2379264
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1036414666760135e-05,
+      "loss": 3.8521,
+      "step": 2379776
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1028028719249616e-05,
+      "loss": 3.8507,
+      "step": 2380288
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1019642771739096e-05,
+      "loss": 3.84,
+      "step": 2380800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.1011256824228574e-05,
+      "loss": 3.8522,
+      "step": 2381312
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.1002903634325518e-05,
+      "loss": 3.8462,
+      "step": 2381824
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0994517686814998e-05,
+      "loss": 3.859,
+      "step": 2382336
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0986131739304478e-05,
+      "loss": 3.8581,
+      "step": 2382848
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0977745791793958e-05,
+      "loss": 3.8459,
+      "step": 2383360
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0969376223087168e-05,
+      "loss": 3.8499,
+      "step": 2383872
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0960990275576648e-05,
+      "loss": 3.8647,
+      "step": 2384384
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.095260432806613e-05,
+      "loss": 3.8459,
+      "step": 2384896
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0944218380555608e-05,
+      "loss": 3.8502,
+      "step": 2385408
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0935832433045088e-05,
+      "loss": 3.849,
+      "step": 2385920
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.092744648553457e-05,
+      "loss": 3.8486,
+      "step": 2386432
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0919060538024048e-05,
+      "loss": 3.848,
+      "step": 2386944
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0910674590513528e-05,
+      "loss": 3.8663,
+      "step": 2387456
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0902288643003008e-05,
+      "loss": 3.8572,
+      "step": 2387968
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.089391907429622e-05,
+      "loss": 3.8532,
+      "step": 2388480
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0885533126785699e-05,
+      "loss": 3.8558,
+      "step": 2388992
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0877147179275179e-05,
+      "loss": 3.8467,
+      "step": 2389504
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.086876123176466e-05,
+      "loss": 3.8626,
+      "step": 2390016
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0860375284254139e-05,
+      "loss": 3.8318,
+      "step": 2390528
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0851989336743619e-05,
+      "loss": 3.8552,
+      "step": 2391040
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0843619768036831e-05,
+      "loss": 3.8577,
+      "step": 2391552
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0835233820526311e-05,
+      "loss": 3.8601,
+      "step": 2392064
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.082684787301579e-05,
+      "loss": 3.8499,
+      "step": 2392576
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0818461925505271e-05,
+      "loss": 3.8398,
+      "step": 2393088
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0810075977994751e-05,
+      "loss": 3.8507,
+      "step": 2393600
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0801706409287962e-05,
+      "loss": 3.8397,
+      "step": 2394112
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0793320461777442e-05,
+      "loss": 3.8674,
+      "step": 2394624
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0784934514266922e-05,
+      "loss": 3.8494,
+      "step": 2395136
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0776548566756402e-05,
+      "loss": 3.8675,
+      "step": 2395648
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0768162619245882e-05,
+      "loss": 3.8501,
+      "step": 2396160
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0759793050539092e-05,
+      "loss": 3.8414,
+      "step": 2396672
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0751423481832305e-05,
+      "loss": 3.8588,
+      "step": 2397184
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0743037534321785e-05,
+      "loss": 3.844,
+      "step": 2397696
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0734651586811263e-05,
+      "loss": 3.8496,
+      "step": 2398208
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0726265639300743e-05,
+      "loss": 3.8475,
+      "step": 2398720
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0717879691790225e-05,
+      "loss": 3.8614,
+      "step": 2399232
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0709493744279703e-05,
+      "loss": 3.8407,
+      "step": 2399744
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0701107796769183e-05,
+      "loss": 3.846,
+      "step": 2400256
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0692721849258663e-05,
+      "loss": 3.8427,
+      "step": 2400768
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0684335901748143e-05,
+      "loss": 3.8407,
+      "step": 2401280
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0675949954237623e-05,
+      "loss": 3.8602,
+      "step": 2401792
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0667580385530833e-05,
+      "loss": 3.8577,
+      "step": 2402304
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0659194438020315e-05,
+      "loss": 3.8462,
+      "step": 2402816
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0650808490509793e-05,
+      "loss": 3.8506,
+      "step": 2403328
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0642422542999273e-05,
+      "loss": 3.8598,
+      "step": 2403840
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0634036595488753e-05,
+      "loss": 3.8545,
+      "step": 2404352
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0625667026781966e-05,
+      "loss": 3.8548,
+      "step": 2404864
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0617281079271444e-05,
+      "loss": 3.8535,
+      "step": 2405376
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0608895131760926e-05,
+      "loss": 3.8521,
+      "step": 2405888
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0600509184250406e-05,
+      "loss": 3.8574,
+      "step": 2406400
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0592123236739884e-05,
+      "loss": 3.8528,
+      "step": 2406912
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0583737289229364e-05,
+      "loss": 3.8504,
+      "step": 2407424
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0575351341718846e-05,
+      "loss": 3.8626,
+      "step": 2407936
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0566965394208324e-05,
+      "loss": 3.8517,
+      "step": 2408448
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0558595825501536e-05,
+      "loss": 3.8373,
+      "step": 2408960
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0550209877991016e-05,
+      "loss": 3.8541,
+      "step": 2409472
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0541823930480496e-05,
+      "loss": 3.8554,
+      "step": 2409984
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0533437982969974e-05,
+      "loss": 3.8529,
+      "step": 2410496
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0525068414263187e-05,
+      "loss": 3.8507,
+      "step": 2411008
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0516682466752667e-05,
+      "loss": 3.8453,
+      "step": 2411520
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.050831289804588e-05,
+      "loss": 3.8519,
+      "step": 2412032
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0499926950535358e-05,
+      "loss": 3.8482,
+      "step": 2412544
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0491541003024838e-05,
+      "loss": 3.8477,
+      "step": 2413056
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0483155055514318e-05,
+      "loss": 3.8539,
+      "step": 2413568
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0474769108003797e-05,
+      "loss": 3.8505,
+      "step": 2414080
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0466399539297008e-05,
+      "loss": 3.8648,
+      "step": 2414592
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.045801359178649e-05,
+      "loss": 3.8355,
+      "step": 2415104
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.044962764427597e-05,
+      "loss": 3.8432,
+      "step": 2415616
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0441241696765448e-05,
+      "loss": 3.8505,
+      "step": 2416128
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0432855749254928e-05,
+      "loss": 3.8486,
+      "step": 2416640
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.042446980174441e-05,
+      "loss": 3.857,
+      "step": 2417152
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0416083854233888e-05,
+      "loss": 3.8544,
+      "step": 2417664
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0407714285527099e-05,
+      "loss": 3.8439,
+      "step": 2418176
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.039932833801658e-05,
+      "loss": 3.8483,
+      "step": 2418688
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.039094239050606e-05,
+      "loss": 3.8513,
+      "step": 2419200
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0382556442995539e-05,
+      "loss": 3.8416,
+      "step": 2419712
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0374186874288751e-05,
+      "loss": 3.8432,
+      "step": 2420224
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0365800926778231e-05,
+      "loss": 3.8526,
+      "step": 2420736
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0357414979267711e-05,
+      "loss": 3.8535,
+      "step": 2421248
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0349029031757191e-05,
+      "loss": 3.8382,
+      "step": 2421760
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0340643084246671e-05,
+      "loss": 3.8446,
+      "step": 2422272
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0332257136736151e-05,
+      "loss": 3.8371,
+      "step": 2422784
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.032387118922563e-05,
+      "loss": 3.8515,
+      "step": 2423296
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.031548524171511e-05,
+      "loss": 3.8426,
+      "step": 2423808
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0307115673008322e-05,
+      "loss": 3.8475,
+      "step": 2424320
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0298729725497802e-05,
+      "loss": 3.8527,
+      "step": 2424832
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0290360156791012e-05,
+      "loss": 3.8523,
+      "step": 2425344
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0281990588084225e-05,
+      "loss": 3.8492,
+      "step": 2425856
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0273604640573705e-05,
+      "loss": 3.8475,
+      "step": 2426368
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0265218693063185e-05,
+      "loss": 3.8541,
+      "step": 2426880
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0256832745552663e-05,
+      "loss": 3.8467,
+      "step": 2427392
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0248446798042145e-05,
+      "loss": 3.8408,
+      "step": 2427904
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0240060850531625e-05,
+      "loss": 3.8526,
+      "step": 2428416
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0231674903021103e-05,
+      "loss": 3.8462,
+      "step": 2428928
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0223288955510583e-05,
+      "loss": 3.8582,
+      "step": 2429440
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0214919386803795e-05,
+      "loss": 3.8433,
+      "step": 2429952
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0206533439293275e-05,
+      "loss": 3.848,
+      "step": 2430464
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0198147491782754e-05,
+      "loss": 3.8436,
+      "step": 2430976
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0189761544272235e-05,
+      "loss": 3.847,
+      "step": 2431488
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0181375596761715e-05,
+      "loss": 3.8377,
+      "step": 2432000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0172989649251193e-05,
+      "loss": 3.8571,
+      "step": 2432512
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0164603701740673e-05,
+      "loss": 3.8638,
+      "step": 2433024
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0156217754230155e-05,
+      "loss": 3.8485,
+      "step": 2433536
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0147848185523366e-05,
+      "loss": 3.8449,
+      "step": 2434048
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0139462238012846e-05,
+      "loss": 3.855,
+      "step": 2434560
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0131076290502326e-05,
+      "loss": 3.8502,
+      "step": 2435072
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0122690342991806e-05,
+      "loss": 3.8455,
+      "step": 2435584
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0114320774285016e-05,
+      "loss": 3.8512,
+      "step": 2436096
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0105934826774496e-05,
+      "loss": 3.8597,
+      "step": 2436608
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0097548879263976e-05,
+      "loss": 3.8544,
+      "step": 2437120
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0089162931753456e-05,
+      "loss": 3.8511,
+      "step": 2437632
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0080776984242936e-05,
+      "loss": 3.8424,
+      "step": 2438144
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0072391036732416e-05,
+      "loss": 3.8555,
+      "step": 2438656
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0064005089221896e-05,
+      "loss": 3.8526,
+      "step": 2439168
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0055651899318838e-05,
+      "loss": 3.8404,
+      "step": 2439680
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0047265951808318e-05,
+      "loss": 3.8641,
+      "step": 2440192
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.00388800042978e-05,
+      "loss": 3.8462,
+      "step": 2440704
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.003049405678728e-05,
+      "loss": 3.8546,
+      "step": 2441216
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0022108109276758e-05,
+      "loss": 3.8554,
+      "step": 2441728
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.0013722161766238e-05,
+      "loss": 3.846,
+      "step": 2442240
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.029672145843506,
+      "eval_runtime": 279.4913,
+      "eval_samples_per_second": 1365.306,
+      "eval_steps_per_second": 42.667,
+      "step": 2442240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.000533621425572e-05,
+      "loss": 3.8476,
+      "step": 2442752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.996950266745198e-06,
+      "loss": 3.8487,
+      "step": 2443264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.988564319234677e-06,
+      "loss": 3.8608,
+      "step": 2443776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.980178371724157e-06,
+      "loss": 3.8503,
+      "step": 2444288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.971792424213637e-06,
+      "loss": 3.8562,
+      "step": 2444800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.963406476703117e-06,
+      "loss": 3.8448,
+      "step": 2445312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.955020529192597e-06,
+      "loss": 3.8469,
+      "step": 2445824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.946634581682077e-06,
+      "loss": 3.8477,
+      "step": 2446336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.938248634171557e-06,
+      "loss": 3.857,
+      "step": 2446848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.929862686661037e-06,
+      "loss": 3.8454,
+      "step": 2447360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.921476739150517e-06,
+      "loss": 3.8591,
+      "step": 2447872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.913107170443728e-06,
+      "loss": 3.8601,
+      "step": 2448384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.904721222933208e-06,
+      "loss": 3.8426,
+      "step": 2448896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.896335275422688e-06,
+      "loss": 3.8466,
+      "step": 2449408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.88794932791217e-06,
+      "loss": 3.8371,
+      "step": 2449920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.879563380401648e-06,
+      "loss": 3.8418,
+      "step": 2450432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.871177432891128e-06,
+      "loss": 3.8507,
+      "step": 2450944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.862791485380608e-06,
+      "loss": 3.8413,
+      "step": 2451456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.854405537870088e-06,
+      "loss": 3.8484,
+      "step": 2451968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.846019590359568e-06,
+      "loss": 3.8692,
+      "step": 2452480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.837633642849048e-06,
+      "loss": 3.8464,
+      "step": 2452992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.829247695338527e-06,
+      "loss": 3.854,
+      "step": 2453504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.820861747828007e-06,
+      "loss": 3.8526,
+      "step": 2454016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.812492179121218e-06,
+      "loss": 3.8472,
+      "step": 2454528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.8041062316107e-06,
+      "loss": 3.8506,
+      "step": 2455040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.795720284100178e-06,
+      "loss": 3.8375,
+      "step": 2455552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.787334336589658e-06,
+      "loss": 3.8481,
+      "step": 2456064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.778948389079138e-06,
+      "loss": 3.8501,
+      "step": 2456576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.770562441568618e-06,
+      "loss": 3.8361,
+      "step": 2457088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.762192872861829e-06,
+      "loss": 3.8486,
+      "step": 2457600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.753806925351309e-06,
+      "loss": 3.8455,
+      "step": 2458112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.74542097784079e-06,
+      "loss": 3.8553,
+      "step": 2458624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.737035030330269e-06,
+      "loss": 3.8525,
+      "step": 2459136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.728649082819749e-06,
+      "loss": 3.8424,
+      "step": 2459648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.720279514112961e-06,
+      "loss": 3.8491,
+      "step": 2460160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.711893566602441e-06,
+      "loss": 3.8638,
+      "step": 2460672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.70350761909192e-06,
+      "loss": 3.8406,
+      "step": 2461184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.695121671581401e-06,
+      "loss": 3.8458,
+      "step": 2461696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.686735724070881e-06,
+      "loss": 3.8472,
+      "step": 2462208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.67834977656036e-06,
+      "loss": 3.8394,
+      "step": 2462720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.669963829049839e-06,
+      "loss": 3.8462,
+      "step": 2463232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.661594260343052e-06,
+      "loss": 3.8626,
+      "step": 2463744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.653208312832532e-06,
+      "loss": 3.8557,
+      "step": 2464256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.644822365322012e-06,
+      "loss": 3.8461,
+      "step": 2464768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.636436417811491e-06,
+      "loss": 3.8541,
+      "step": 2465280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.628050470300971e-06,
+      "loss": 3.8439,
+      "step": 2465792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.619680901594182e-06,
+      "loss": 3.8562,
+      "step": 2466304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.611294954083662e-06,
+      "loss": 3.8312,
+      "step": 2466816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.602909006573142e-06,
+      "loss": 3.8474,
+      "step": 2467328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.594523059062622e-06,
+      "loss": 3.8567,
+      "step": 2467840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.586137111552102e-06,
+      "loss": 3.8512,
+      "step": 2468352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.577751164041582e-06,
+      "loss": 3.8504,
+      "step": 2468864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.569365216531062e-06,
+      "loss": 3.84,
+      "step": 2469376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.56097926902054e-06,
+      "loss": 3.848,
+      "step": 2469888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.552609700313753e-06,
+      "loss": 3.8364,
+      "step": 2470400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.544223752803233e-06,
+      "loss": 3.862,
+      "step": 2470912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.535854184096445e-06,
+      "loss": 3.8456,
+      "step": 2471424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.527468236585923e-06,
+      "loss": 3.8594,
+      "step": 2471936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.519082289075403e-06,
+      "loss": 3.8472,
+      "step": 2472448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.510696341564885e-06,
+      "loss": 3.8419,
+      "step": 2472960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.502310394054363e-06,
+      "loss": 3.8515,
+      "step": 2473472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.493924446543843e-06,
+      "loss": 3.8446,
+      "step": 2473984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.485538499033323e-06,
+      "loss": 3.8472,
+      "step": 2474496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.477168930326536e-06,
+      "loss": 3.8426,
+      "step": 2475008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.468782982816014e-06,
+      "loss": 3.8573,
+      "step": 2475520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.460397035305494e-06,
+      "loss": 3.841,
+      "step": 2476032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.452027466598706e-06,
+      "loss": 3.8429,
+      "step": 2476544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.443641519088185e-06,
+      "loss": 3.8366,
+      "step": 2477056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.435255571577666e-06,
+      "loss": 3.8389,
+      "step": 2477568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.426869624067146e-06,
+      "loss": 3.8605,
+      "step": 2478080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.418483676556626e-06,
+      "loss": 3.8568,
+      "step": 2478592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.410097729046104e-06,
+      "loss": 3.8474,
+      "step": 2479104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.401711781535586e-06,
+      "loss": 3.8467,
+      "step": 2479616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.393325834025066e-06,
+      "loss": 3.8539,
+      "step": 2480128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.384939886514544e-06,
+      "loss": 3.854,
+      "step": 2480640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.376553939004024e-06,
+      "loss": 3.847,
+      "step": 2481152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.368167991493506e-06,
+      "loss": 3.8537,
+      "step": 2481664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.359782043982984e-06,
+      "loss": 3.8501,
+      "step": 2482176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.351412475276197e-06,
+      "loss": 3.8527,
+      "step": 2482688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.343026527765677e-06,
+      "loss": 3.8473,
+      "step": 2483200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.334640580255157e-06,
+      "loss": 3.851,
+      "step": 2483712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.326254632744635e-06,
+      "loss": 3.8588,
+      "step": 2484224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.317868685234117e-06,
+      "loss": 3.8477,
+      "step": 2484736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.309499116527327e-06,
+      "loss": 3.834,
+      "step": 2485248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.301113169016807e-06,
+      "loss": 3.8493,
+      "step": 2485760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.292727221506287e-06,
+      "loss": 3.8497,
+      "step": 2486272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.284341273995767e-06,
+      "loss": 3.8524,
+      "step": 2486784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.275955326485247e-06,
+      "loss": 3.8482,
+      "step": 2487296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.267585757778458e-06,
+      "loss": 3.8415,
+      "step": 2487808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.259199810267938e-06,
+      "loss": 3.8506,
+      "step": 2488320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.250813862757418e-06,
+      "loss": 3.8412,
+      "step": 2488832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.242427915246898e-06,
+      "loss": 3.8477,
+      "step": 2489344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.234041967736378e-06,
+      "loss": 3.8464,
+      "step": 2489856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.225656020225858e-06,
+      "loss": 3.8444,
+      "step": 2490368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.21728645151907e-06,
+      "loss": 3.8641,
+      "step": 2490880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.208900504008548e-06,
+      "loss": 3.8342,
+      "step": 2491392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.200514556498028e-06,
+      "loss": 3.8353,
+      "step": 2491904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.192128608987508e-06,
+      "loss": 3.8512,
+      "step": 2492416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.18374266147699e-06,
+      "loss": 3.8487,
+      "step": 2492928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.175356713966468e-06,
+      "loss": 3.8568,
+      "step": 2493440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.166970766455948e-06,
+      "loss": 3.8477,
+      "step": 2493952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.158584818945428e-06,
+      "loss": 3.8404,
+      "step": 2494464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.150215250238639e-06,
+      "loss": 3.8446,
+      "step": 2494976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.141829302728119e-06,
+      "loss": 3.8517,
+      "step": 2495488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.133443355217599e-06,
+      "loss": 3.8383,
+      "step": 2496000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.125073786510811e-06,
+      "loss": 3.8367,
+      "step": 2496512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.11668783900029e-06,
+      "loss": 3.8512,
+      "step": 2497024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.108301891489771e-06,
+      "loss": 3.8464,
+      "step": 2497536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.099915943979251e-06,
+      "loss": 3.8358,
+      "step": 2498048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.09152999646873e-06,
+      "loss": 3.839,
+      "step": 2498560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.08314404895821e-06,
+      "loss": 3.8395,
+      "step": 2499072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.074758101447691e-06,
+      "loss": 3.8498,
+      "step": 2499584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.066372153937171e-06,
+      "loss": 3.8392,
+      "step": 2500096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.058018964034113e-06,
+      "loss": 3.8462,
+      "step": 2500608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.049633016523593e-06,
+      "loss": 3.8484,
+      "step": 2501120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.041247069013073e-06,
+      "loss": 3.8523,
+      "step": 2501632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.032861121502553e-06,
+      "loss": 3.8408,
+      "step": 2502144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.024475173992032e-06,
+      "loss": 3.8478,
+      "step": 2502656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.016089226481512e-06,
+      "loss": 3.8489,
+      "step": 2503168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.007703278970992e-06,
+      "loss": 3.8453,
+      "step": 2503680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.999317331460472e-06,
+      "loss": 3.8321,
+      "step": 2504192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.990947762753683e-06,
+      "loss": 3.8526,
+      "step": 2504704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.982561815243163e-06,
+      "loss": 3.8436,
+      "step": 2505216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.974175867732643e-06,
+      "loss": 3.8521,
+      "step": 2505728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.965789920222123e-06,
+      "loss": 3.8384,
+      "step": 2506240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.957403972711603e-06,
+      "loss": 3.85,
+      "step": 2506752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.949034404004815e-06,
+      "loss": 3.8419,
+      "step": 2507264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.940664835298025e-06,
+      "loss": 3.8437,
+      "step": 2507776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.932278887787506e-06,
+      "loss": 3.8301,
+      "step": 2508288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.923892940276986e-06,
+      "loss": 3.8595,
+      "step": 2508800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.915506992766466e-06,
+      "loss": 3.8594,
+      "step": 2509312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.907121045255944e-06,
+      "loss": 3.846,
+      "step": 2509824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.898751476549157e-06,
+      "loss": 3.8371,
+      "step": 2510336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.890365529038637e-06,
+      "loss": 3.8603,
+      "step": 2510848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.881979581528117e-06,
+      "loss": 3.8463,
+      "step": 2511360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.873593634017597e-06,
+      "loss": 3.8456,
+      "step": 2511872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.865224065310808e-06,
+      "loss": 3.8466,
+      "step": 2512384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.856838117800287e-06,
+      "loss": 3.8547,
+      "step": 2512896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.848452170289767e-06,
+      "loss": 3.8474,
+      "step": 2513408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.840066222779247e-06,
+      "loss": 3.8498,
+      "step": 2513920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.831680275268727e-06,
+      "loss": 3.8354,
+      "step": 2514432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.823294327758207e-06,
+      "loss": 3.8534,
+      "step": 2514944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.814908380247687e-06,
+      "loss": 3.8511,
+      "step": 2515456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.806522432737167e-06,
+      "loss": 3.8407,
+      "step": 2515968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.79815286403038e-06,
+      "loss": 3.8613,
+      "step": 2516480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.789766916519858e-06,
+      "loss": 3.8419,
+      "step": 2516992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.781380969009338e-06,
+      "loss": 3.8526,
+      "step": 2517504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.772995021498818e-06,
+      "loss": 3.8524,
+      "step": 2518016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.764609073988298e-06,
+      "loss": 3.8448,
+      "step": 2518528
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.028461933135986,
+      "eval_runtime": 297.4393,
+      "eval_samples_per_second": 1282.921,
+      "eval_steps_per_second": 40.092,
+      "step": 2518560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.756223126477778e-06,
+      "loss": 3.8394,
+      "step": 2519040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.747837178967258e-06,
+      "loss": 3.8491,
+      "step": 2519552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.739451231456738e-06,
+      "loss": 3.8558,
+      "step": 2520064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.731065283946218e-06,
+      "loss": 3.8492,
+      "step": 2520576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.722679336435698e-06,
+      "loss": 3.8574,
+      "step": 2521088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.714293388925178e-06,
+      "loss": 3.842,
+      "step": 2521600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.705907441414658e-06,
+      "loss": 3.8426,
+      "step": 2522112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.697521493904137e-06,
+      "loss": 3.8436,
+      "step": 2522624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.689135546393617e-06,
+      "loss": 3.8552,
+      "step": 2523136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.680749598883096e-06,
+      "loss": 3.843,
+      "step": 2523648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.672363651372577e-06,
+      "loss": 3.8519,
+      "step": 2524160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.663977703862057e-06,
+      "loss": 3.8552,
+      "step": 2524672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.655608135155268e-06,
+      "loss": 3.8421,
+      "step": 2525184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.647222187644748e-06,
+      "loss": 3.8417,
+      "step": 2525696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.638836240134228e-06,
+      "loss": 3.8348,
+      "step": 2526208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.630450292623708e-06,
+      "loss": 3.8426,
+      "step": 2526720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.622064345113188e-06,
+      "loss": 3.8437,
+      "step": 2527232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.613678397602668e-06,
+      "loss": 3.8402,
+      "step": 2527744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.605292450092148e-06,
+      "loss": 3.8445,
+      "step": 2528256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.596906502581628e-06,
+      "loss": 3.8662,
+      "step": 2528768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.588520555071108e-06,
+      "loss": 3.8434,
+      "step": 2529280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.580134607560588e-06,
+      "loss": 3.848,
+      "step": 2529792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.571765038853798e-06,
+      "loss": 3.8556,
+      "step": 2530304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.563379091343278e-06,
+      "loss": 3.8474,
+      "step": 2530816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.554993143832758e-06,
+      "loss": 3.8417,
+      "step": 2531328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.546607196322238e-06,
+      "loss": 3.8366,
+      "step": 2531840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.538221248811718e-06,
+      "loss": 3.8437,
+      "step": 2532352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.529851680104929e-06,
+      "loss": 3.8449,
+      "step": 2532864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.521465732594409e-06,
+      "loss": 3.8327,
+      "step": 2533376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.513079785083889e-06,
+      "loss": 3.8465,
+      "step": 2533888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.504693837573369e-06,
+      "loss": 3.839,
+      "step": 2534400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.496307890062849e-06,
+      "loss": 3.8532,
+      "step": 2534912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.487938321356061e-06,
+      "loss": 3.8515,
+      "step": 2535424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.479552373845541e-06,
+      "loss": 3.8362,
+      "step": 2535936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.47118280513875e-06,
+      "loss": 3.8516,
+      "step": 2536448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.462796857628232e-06,
+      "loss": 3.8568,
+      "step": 2536960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.454410910117712e-06,
+      "loss": 3.8354,
+      "step": 2537472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.446024962607192e-06,
+      "loss": 3.8478,
+      "step": 2537984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.43763901509667e-06,
+      "loss": 3.8413,
+      "step": 2538496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.429253067586152e-06,
+      "loss": 3.8366,
+      "step": 2539008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.420867120075632e-06,
+      "loss": 3.8446,
+      "step": 2539520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.41248117256511e-06,
+      "loss": 3.8575,
+      "step": 2540032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.404095225054592e-06,
+      "loss": 3.8502,
+      "step": 2540544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.395709277544072e-06,
+      "loss": 3.8445,
+      "step": 2541056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.38732333003355e-06,
+      "loss": 3.8526,
+      "step": 2541568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.378953761326762e-06,
+      "loss": 3.8404,
+      "step": 2542080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.370584192619973e-06,
+      "loss": 3.8519,
+      "step": 2542592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.362198245109453e-06,
+      "loss": 3.8306,
+      "step": 2543104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.353812297598933e-06,
+      "loss": 3.8411,
+      "step": 2543616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.345426350088413e-06,
+      "loss": 3.8543,
+      "step": 2544128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.337040402577893e-06,
+      "loss": 3.8522,
+      "step": 2544640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.328654455067373e-06,
+      "loss": 3.8449,
+      "step": 2545152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.320268507556853e-06,
+      "loss": 3.8342,
+      "step": 2545664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.311882560046333e-06,
+      "loss": 3.8483,
+      "step": 2546176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.303512991339544e-06,
+      "loss": 3.83,
+      "step": 2546688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.295143422632755e-06,
+      "loss": 3.8578,
+      "step": 2547200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.286757475122234e-06,
+      "loss": 3.8454,
+      "step": 2547712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.278371527611716e-06,
+      "loss": 3.8594,
+      "step": 2548224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.269985580101196e-06,
+      "loss": 3.8453,
+      "step": 2548736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.261599632590674e-06,
+      "loss": 3.8398,
+      "step": 2549248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.253213685080154e-06,
+      "loss": 3.8427,
+      "step": 2549760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.244827737569636e-06,
+      "loss": 3.8484,
+      "step": 2550272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.236441790059114e-06,
+      "loss": 3.8374,
+      "step": 2550784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.228072221352327e-06,
+      "loss": 3.837,
+      "step": 2551296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.219686273841807e-06,
+      "loss": 3.8549,
+      "step": 2551808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.211300326331287e-06,
+      "loss": 3.8402,
+      "step": 2552320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.202914378820765e-06,
+      "loss": 3.8339,
+      "step": 2552832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.194528431310247e-06,
+      "loss": 3.8346,
+      "step": 2553344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.186142483799726e-06,
+      "loss": 3.8351,
+      "step": 2553856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.177772915092936e-06,
+      "loss": 3.8529,
+      "step": 2554368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.169386967582417e-06,
+      "loss": 3.8546,
+      "step": 2554880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.161001020071897e-06,
+      "loss": 3.8408,
+      "step": 2555392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.152631451365108e-06,
+      "loss": 3.8455,
+      "step": 2555904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.144245503854588e-06,
+      "loss": 3.8504,
+      "step": 2556416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.135859556344068e-06,
+      "loss": 3.8486,
+      "step": 2556928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.127473608833548e-06,
+      "loss": 3.8439,
+      "step": 2557440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.119087661323028e-06,
+      "loss": 3.8517,
+      "step": 2557952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.110701713812508e-06,
+      "loss": 3.847,
+      "step": 2558464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.102315766301988e-06,
+      "loss": 3.8537,
+      "step": 2558976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.093929818791468e-06,
+      "loss": 3.8425,
+      "step": 2559488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.085560250084678e-06,
+      "loss": 3.8486,
+      "step": 2560000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.077174302574158e-06,
+      "loss": 3.852,
+      "step": 2560512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.068788355063638e-06,
+      "loss": 3.8492,
+      "step": 2561024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.060402407553118e-06,
+      "loss": 3.8323,
+      "step": 2561536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.052016460042598e-06,
+      "loss": 3.8445,
+      "step": 2562048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.043646891335809e-06,
+      "loss": 3.8436,
+      "step": 2562560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.03526094382529e-06,
+      "loss": 3.8499,
+      "step": 2563072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.026874996314769e-06,
+      "loss": 3.8461,
+      "step": 2563584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.018489048804249e-06,
+      "loss": 3.8356,
+      "step": 2564096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.010103101293729e-06,
+      "loss": 3.8458,
+      "step": 2564608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.001733532586941e-06,
+      "loss": 3.8387,
+      "step": 2565120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.99334758507642e-06,
+      "loss": 3.8474,
+      "step": 2565632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.984978016369632e-06,
+      "loss": 3.838,
+      "step": 2566144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.976592068859112e-06,
+      "loss": 3.844,
+      "step": 2566656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.96820612134859e-06,
+      "loss": 3.8533,
+      "step": 2567168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.959820173838072e-06,
+      "loss": 3.8452,
+      "step": 2567680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.951434226327552e-06,
+      "loss": 3.8279,
+      "step": 2568192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.943048278817032e-06,
+      "loss": 3.8454,
+      "step": 2568704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.934662331306512e-06,
+      "loss": 3.8463,
+      "step": 2569216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.926276383795992e-06,
+      "loss": 3.8527,
+      "step": 2569728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.917890436285472e-06,
+      "loss": 3.8452,
+      "step": 2570240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.90950448877495e-06,
+      "loss": 3.8406,
+      "step": 2570752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.901118541264432e-06,
+      "loss": 3.8373,
+      "step": 2571264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.892732593753912e-06,
+      "loss": 3.8509,
+      "step": 2571776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.884363025047122e-06,
+      "loss": 3.8302,
+      "step": 2572288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.875993456340333e-06,
+      "loss": 3.8386,
+      "step": 2572800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.867607508829813e-06,
+      "loss": 3.8475,
+      "step": 2573312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.859221561319293e-06,
+      "loss": 3.8472,
+      "step": 2573824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.850835613808773e-06,
+      "loss": 3.828,
+      "step": 2574336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.842449666298253e-06,
+      "loss": 3.8421,
+      "step": 2574848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.834080097591464e-06,
+      "loss": 3.8303,
+      "step": 2575360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.825694150080945e-06,
+      "loss": 3.847,
+      "step": 2575872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.817308202570424e-06,
+      "loss": 3.8378,
+      "step": 2576384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.808922255059904e-06,
+      "loss": 3.8345,
+      "step": 2576896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.800536307549385e-06,
+      "loss": 3.8513,
+      "step": 2577408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.792150360038864e-06,
+      "loss": 3.8464,
+      "step": 2577920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.783780791332074e-06,
+      "loss": 3.8379,
+      "step": 2578432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.775394843821556e-06,
+      "loss": 3.8452,
+      "step": 2578944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.767008896311036e-06,
+      "loss": 3.8423,
+      "step": 2579456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.758622948800514e-06,
+      "loss": 3.8458,
+      "step": 2579968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.750237001289994e-06,
+      "loss": 3.8275,
+      "step": 2580480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.741851053779476e-06,
+      "loss": 3.853,
+      "step": 2580992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.733481485072685e-06,
+      "loss": 3.8375,
+      "step": 2581504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.725095537562167e-06,
+      "loss": 3.8482,
+      "step": 2582016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.716709590051647e-06,
+      "loss": 3.8351,
+      "step": 2582528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.708323642541127e-06,
+      "loss": 3.8442,
+      "step": 2583040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.699937695030605e-06,
+      "loss": 3.8391,
+      "step": 2583552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.691568126323817e-06,
+      "loss": 3.8388,
+      "step": 2584064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.683182178813297e-06,
+      "loss": 3.8249,
+      "step": 2584576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.674796231302775e-06,
+      "loss": 3.8596,
+      "step": 2585088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.666410283792257e-06,
+      "loss": 3.8501,
+      "step": 2585600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.658024336281737e-06,
+      "loss": 3.8457,
+      "step": 2586112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.649654767574948e-06,
+      "loss": 3.8362,
+      "step": 2586624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.641285198868159e-06,
+      "loss": 3.8529,
+      "step": 2587136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.632899251357639e-06,
+      "loss": 3.8492,
+      "step": 2587648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.6245133038471194e-06,
+      "loss": 3.8408,
+      "step": 2588160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.616127356336599e-06,
+      "loss": 3.8404,
+      "step": 2588672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.6077414088260785e-06,
+      "loss": 3.8512,
+      "step": 2589184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.599355461315559e-06,
+      "loss": 3.8445,
+      "step": 2589696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.590969513805039e-06,
+      "loss": 3.8539,
+      "step": 2590208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.582583566294518e-06,
+      "loss": 3.8329,
+      "step": 2590720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.574197618783998e-06,
+      "loss": 3.8501,
+      "step": 2591232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.565811671273479e-06,
+      "loss": 3.8492,
+      "step": 2591744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.557425723762958e-06,
+      "loss": 3.8397,
+      "step": 2592256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.54905615505617e-06,
+      "loss": 3.857,
+      "step": 2592768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.54067020754565e-06,
+      "loss": 3.8403,
+      "step": 2593280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.53228426003513e-06,
+      "loss": 3.845,
+      "step": 2593792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.523898312524609e-06,
+      "loss": 3.8442,
+      "step": 2594304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.51551236501409e-06,
+      "loss": 3.8429,
+      "step": 2594816
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.027493476867676,
+      "eval_runtime": 290.5589,
+      "eval_samples_per_second": 1313.3,
+      "eval_steps_per_second": 41.042,
+      "step": 2594880
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.507142796307301e-06,
+      "loss": 3.8394,
+      "step": 2595328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.498756848796781e-06,
+      "loss": 3.8476,
+      "step": 2595840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.49037090128626e-06,
+      "loss": 3.8518,
+      "step": 2596352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.48198495377574e-06,
+      "loss": 3.8493,
+      "step": 2596864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.473599006265221e-06,
+      "loss": 3.8534,
+      "step": 2597376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.465213058754699e-06,
+      "loss": 3.8359,
+      "step": 2597888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.45682711124418e-06,
+      "loss": 3.8385,
+      "step": 2598400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.448473921341123e-06,
+      "loss": 3.8397,
+      "step": 2598912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.4400879738306035e-06,
+      "loss": 3.8513,
+      "step": 2599424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.431702026320083e-06,
+      "loss": 3.8408,
+      "step": 2599936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.4233160788095626e-06,
+      "loss": 3.8504,
+      "step": 2600448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.414930131299043e-06,
+      "loss": 3.8514,
+      "step": 2600960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.406560562592255e-06,
+      "loss": 3.8374,
+      "step": 2601472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.398174615081733e-06,
+      "loss": 3.8434,
+      "step": 2601984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.389788667571214e-06,
+      "loss": 3.8283,
+      "step": 2602496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.381402720060694e-06,
+      "loss": 3.8388,
+      "step": 2603008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.373016772550173e-06,
+      "loss": 3.842,
+      "step": 2603520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.364630825039653e-06,
+      "loss": 3.8382,
+      "step": 2604032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.356244877529134e-06,
+      "loss": 3.8375,
+      "step": 2604544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.347858930018613e-06,
+      "loss": 3.8625,
+      "step": 2605056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.339472982508093e-06,
+      "loss": 3.8407,
+      "step": 2605568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.331087034997573e-06,
+      "loss": 3.8446,
+      "step": 2606080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.322701087487054e-06,
+      "loss": 3.853,
+      "step": 2606592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.314331518780264e-06,
+      "loss": 3.8471,
+      "step": 2607104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.3059455712697445e-06,
+      "loss": 3.8396,
+      "step": 2607616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.297559623759224e-06,
+      "loss": 3.8325,
+      "step": 2608128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.2891736762487035e-06,
+      "loss": 3.8432,
+      "step": 2608640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.2807877287381835e-06,
+      "loss": 3.8396,
+      "step": 2609152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.272418160031395e-06,
+      "loss": 3.8315,
+      "step": 2609664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.264032212520876e-06,
+      "loss": 3.8455,
+      "step": 2610176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.255646265010355e-06,
+      "loss": 3.8387,
+      "step": 2610688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.247260317499835e-06,
+      "loss": 3.8494,
+      "step": 2611200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.238890748793047e-06,
+      "loss": 3.8466,
+      "step": 2611712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.2305048012825266e-06,
+      "loss": 3.8337,
+      "step": 2612224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.222135232575737e-06,
+      "loss": 3.8465,
+      "step": 2612736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.213749285065217e-06,
+      "loss": 3.8573,
+      "step": 2613248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.205363337554698e-06,
+      "loss": 3.8279,
+      "step": 2613760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.196977390044176e-06,
+      "loss": 3.8463,
+      "step": 2614272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.188591442533657e-06,
+      "loss": 3.8366,
+      "step": 2614784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.180205495023137e-06,
+      "loss": 3.8375,
+      "step": 2615296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.171819547512616e-06,
+      "loss": 3.8397,
+      "step": 2615808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.163449978805828e-06,
+      "loss": 3.8555,
+      "step": 2616320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.155064031295308e-06,
+      "loss": 3.8469,
+      "step": 2616832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.146678083784789e-06,
+      "loss": 3.8421,
+      "step": 2617344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.138292136274268e-06,
+      "loss": 3.8485,
+      "step": 2617856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.129906188763748e-06,
+      "loss": 3.8354,
+      "step": 2618368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.121536620056959e-06,
+      "loss": 3.8509,
+      "step": 2618880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.11315067254644e-06,
+      "loss": 3.8303,
+      "step": 2619392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.104764725035918e-06,
+      "loss": 3.837,
+      "step": 2619904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.096378777525399e-06,
+      "loss": 3.8488,
+      "step": 2620416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.087992830014879e-06,
+      "loss": 3.8469,
+      "step": 2620928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.079606882504358e-06,
+      "loss": 3.8412,
+      "step": 2621440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.071220934993838e-06,
+      "loss": 3.8361,
+      "step": 2621952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.062834987483319e-06,
+      "loss": 3.8441,
+      "step": 2622464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.054449039972798e-06,
+      "loss": 3.8291,
+      "step": 2622976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.046063092462278e-06,
+      "loss": 3.8517,
+      "step": 2623488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.037677144951758e-06,
+      "loss": 3.8455,
+      "step": 2624000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.029291197441239e-06,
+      "loss": 3.8519,
+      "step": 2624512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.020921628734449e-06,
+      "loss": 3.8463,
+      "step": 2625024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.01253568122393e-06,
+      "loss": 3.8349,
+      "step": 2625536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.004166112517141e-06,
+      "loss": 3.8404,
+      "step": 2626048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.995780165006621e-06,
+      "loss": 3.8458,
+      "step": 2626560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.9873942174961e-06,
+      "loss": 3.8337,
+      "step": 2627072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.97900826998558e-06,
+      "loss": 3.8369,
+      "step": 2627584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.970638701278792e-06,
+      "loss": 3.8509,
+      "step": 2628096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.962252753768271e-06,
+      "loss": 3.8413,
+      "step": 2628608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.953866806257751e-06,
+      "loss": 3.8315,
+      "step": 2629120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.9454972375509635e-06,
+      "loss": 3.833,
+      "step": 2629632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.937111290040443e-06,
+      "loss": 3.8293,
+      "step": 2630144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.9287253425299225e-06,
+      "loss": 3.8523,
+      "step": 2630656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.9203393950194025e-06,
+      "loss": 3.8503,
+      "step": 2631168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.911953447508883e-06,
+      "loss": 3.8391,
+      "step": 2631680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.9035674999983615e-06,
+      "loss": 3.8467,
+      "step": 2632192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.895181552487842e-06,
+      "loss": 3.8422,
+      "step": 2632704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.886795604977322e-06,
+      "loss": 3.8498,
+      "step": 2633216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.878409657466803e-06,
+      "loss": 3.8384,
+      "step": 2633728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.870023709956281e-06,
+      "loss": 3.8504,
+      "step": 2634240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.861654141249493e-06,
+      "loss": 3.8467,
+      "step": 2634752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.853268193738974e-06,
+      "loss": 3.8462,
+      "step": 2635264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.844882246228453e-06,
+      "loss": 3.8397,
+      "step": 2635776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.8365126775216645e-06,
+      "loss": 3.8459,
+      "step": 2636288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.8281267300111445e-06,
+      "loss": 3.8476,
+      "step": 2636800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.8197407825006245e-06,
+      "loss": 3.8477,
+      "step": 2637312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.8113548349901036e-06,
+      "loss": 3.8336,
+      "step": 2637824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.802968887479584e-06,
+      "loss": 3.8424,
+      "step": 2638336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.794582939969064e-06,
+      "loss": 3.8407,
+      "step": 2638848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.7861969924585434e-06,
+      "loss": 3.8455,
+      "step": 2639360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.777811044948023e-06,
+      "loss": 3.8489,
+      "step": 2639872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.769425097437504e-06,
+      "loss": 3.8303,
+      "step": 2640384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.761039149926984e-06,
+      "loss": 3.8428,
+      "step": 2640896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.752653202416463e-06,
+      "loss": 3.8388,
+      "step": 2641408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.744267254905943e-06,
+      "loss": 3.8456,
+      "step": 2641920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.735881307395424e-06,
+      "loss": 3.8302,
+      "step": 2642432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.727511738688634e-06,
+      "loss": 3.8426,
+      "step": 2642944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.719125791178115e-06,
+      "loss": 3.8523,
+      "step": 2643456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.710739843667595e-06,
+      "loss": 3.8398,
+      "step": 2643968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.702353896157075e-06,
+      "loss": 3.8243,
+      "step": 2644480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.693967948646554e-06,
+      "loss": 3.8429,
+      "step": 2644992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.685598379939765e-06,
+      "loss": 3.8474,
+      "step": 2645504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.677212432429246e-06,
+      "loss": 3.8474,
+      "step": 2646016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.6688264849187245e-06,
+      "loss": 3.8426,
+      "step": 2646528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.660440537408205e-06,
+      "loss": 3.8351,
+      "step": 2647040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.652054589897685e-06,
+      "loss": 3.8361,
+      "step": 2647552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.643668642387164e-06,
+      "loss": 3.8475,
+      "step": 2648064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.635282694876645e-06,
+      "loss": 3.8298,
+      "step": 2648576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.626913126169857e-06,
+      "loss": 3.8364,
+      "step": 2649088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.618543557463068e-06,
+      "loss": 3.8443,
+      "step": 2649600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.610157609952547e-06,
+      "loss": 3.8445,
+      "step": 2650112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.6017716624420275e-06,
+      "loss": 3.8226,
+      "step": 2650624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.5933857149315074e-06,
+      "loss": 3.8409,
+      "step": 2651136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.584999767420988e-06,
+      "loss": 3.829,
+      "step": 2651648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.5766138199104665e-06,
+      "loss": 3.8414,
+      "step": 2652160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.568227872399947e-06,
+      "loss": 3.8378,
+      "step": 2652672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.559841924889427e-06,
+      "loss": 3.8326,
+      "step": 2653184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.551455977378906e-06,
+      "loss": 3.8496,
+      "step": 2653696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.543070029868386e-06,
+      "loss": 3.8457,
+      "step": 2654208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.534700461161598e-06,
+      "loss": 3.8344,
+      "step": 2654720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.526314513651079e-06,
+      "loss": 3.8384,
+      "step": 2655232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.517928566140558e-06,
+      "loss": 3.8412,
+      "step": 2655744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.509542618630038e-06,
+      "loss": 3.8401,
+      "step": 2656256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.5011730499232495e-06,
+      "loss": 3.8278,
+      "step": 2656768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.492787102412729e-06,
+      "loss": 3.8474,
+      "step": 2657280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.4844011549022085e-06,
+      "loss": 3.8382,
+      "step": 2657792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.476015207391689e-06,
+      "loss": 3.8448,
+      "step": 2658304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.467629259881169e-06,
+      "loss": 3.8374,
+      "step": 2658816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.459243312370648e-06,
+      "loss": 3.8453,
+      "step": 2659328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.450857364860128e-06,
+      "loss": 3.8307,
+      "step": 2659840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.44248779615334e-06,
+      "loss": 3.837,
+      "step": 2660352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.434101848642819e-06,
+      "loss": 3.826,
+      "step": 2660864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.425732279936031e-06,
+      "loss": 3.8534,
+      "step": 2661376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.4173463324255115e-06,
+      "loss": 3.8463,
+      "step": 2661888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.4089603849149915e-06,
+      "loss": 3.8448,
+      "step": 2662400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.400574437404471e-06,
+      "loss": 3.8326,
+      "step": 2662912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.3921884898939506e-06,
+      "loss": 3.8501,
+      "step": 2663424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.383802542383431e-06,
+      "loss": 3.8469,
+      "step": 2663936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.37541659487291e-06,
+      "loss": 3.8356,
+      "step": 2664448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.3670306473623904e-06,
+      "loss": 3.8398,
+      "step": 2664960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.35864469985187e-06,
+      "loss": 3.8467,
+      "step": 2665472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.350258752341351e-06,
+      "loss": 3.8412,
+      "step": 2665984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.341889183634561e-06,
+      "loss": 3.8502,
+      "step": 2666496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.333503236124041e-06,
+      "loss": 3.8327,
+      "step": 2667008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.325117288613522e-06,
+      "loss": 3.8459,
+      "step": 2667520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.316731341103001e-06,
+      "loss": 3.8406,
+      "step": 2668032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.308345393592481e-06,
+      "loss": 3.8345,
+      "step": 2668544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.299975824885693e-06,
+      "loss": 3.8558,
+      "step": 2669056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.291589877375173e-06,
+      "loss": 3.8376,
+      "step": 2669568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.283203929864652e-06,
+      "loss": 3.8423,
+      "step": 2670080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.2748179823541325e-06,
+      "loss": 3.8449,
+      "step": 2670592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.266432034843612e-06,
+      "loss": 3.8383,
+      "step": 2671104
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.027126789093018,
+      "eval_runtime": 289.9721,
+      "eval_samples_per_second": 1315.958,
+      "eval_steps_per_second": 41.125,
+      "step": 2671200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.2580460873330915e-06,
+      "loss": 3.8366,
+      "step": 2671616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.2496601398225715e-06,
+      "loss": 3.8414,
+      "step": 2672128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.241274192312052e-06,
+      "loss": 3.8475,
+      "step": 2672640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.232888244801531e-06,
+      "loss": 3.8443,
+      "step": 2673152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.224502297291011e-06,
+      "loss": 3.8513,
+      "step": 2673664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.216116349780491e-06,
+      "loss": 3.8359,
+      "step": 2674176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.207730402269971e-06,
+      "loss": 3.84,
+      "step": 2674688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.199344454759451e-06,
+      "loss": 3.832,
+      "step": 2675200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.190958507248931e-06,
+      "loss": 3.8524,
+      "step": 2675712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.182572559738411e-06,
+      "loss": 3.8376,
+      "step": 2676224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.174186612227891e-06,
+      "loss": 3.8486,
+      "step": 2676736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.165800664717371e-06,
+      "loss": 3.8473,
+      "step": 2677248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.157431096010583e-06,
+      "loss": 3.841,
+      "step": 2677760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.149045148500062e-06,
+      "loss": 3.8309,
+      "step": 2678272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.140659200989543e-06,
+      "loss": 3.8317,
+      "step": 2678784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.132273253479022e-06,
+      "loss": 3.833,
+      "step": 2679296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.123887305968502e-06,
+      "loss": 3.841,
+      "step": 2679808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.115501358457982e-06,
+      "loss": 3.8354,
+      "step": 2680320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.1071154109474616e-06,
+      "loss": 3.8372,
+      "step": 2680832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.0987294634369415e-06,
+      "loss": 3.8585,
+      "step": 2681344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.0903435159264215e-06,
+      "loss": 3.8413,
+      "step": 2681856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.081973947219633e-06,
+      "loss": 3.8393,
+      "step": 2682368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.073587999709113e-06,
+      "loss": 3.8501,
+      "step": 2682880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.065202052198592e-06,
+      "loss": 3.8443,
+      "step": 2683392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.056816104688073e-06,
+      "loss": 3.8363,
+      "step": 2683904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.048446535981284e-06,
+      "loss": 3.8305,
+      "step": 2684416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.040060588470764e-06,
+      "loss": 3.8456,
+      "step": 2684928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.031674640960244e-06,
+      "loss": 3.8315,
+      "step": 2685440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.023288693449724e-06,
+      "loss": 3.8325,
+      "step": 2685952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.014902745939204e-06,
+      "loss": 3.8368,
+      "step": 2686464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.0065167984286835e-06,
+      "loss": 3.834,
+      "step": 2686976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9981308509181635e-06,
+      "loss": 3.8479,
+      "step": 2687488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9897449034076435e-06,
+      "loss": 3.8422,
+      "step": 2688000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.981375334700855e-06,
+      "loss": 3.8332,
+      "step": 2688512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.973005765994066e-06,
+      "loss": 3.8454,
+      "step": 2689024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9646361972872775e-06,
+      "loss": 3.8507,
+      "step": 2689536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.956250249776757e-06,
+      "loss": 3.8313,
+      "step": 2690048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9478643022662374e-06,
+      "loss": 3.8433,
+      "step": 2690560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9394783547557165e-06,
+      "loss": 3.8295,
+      "step": 2691072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.931092407245197e-06,
+      "loss": 3.8421,
+      "step": 2691584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9227064597346764e-06,
+      "loss": 3.8334,
+      "step": 2692096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.914320512224156e-06,
+      "loss": 3.853,
+      "step": 2692608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.905934564713636e-06,
+      "loss": 3.8424,
+      "step": 2693120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.897548617203116e-06,
+      "loss": 3.8412,
+      "step": 2693632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.889162669692596e-06,
+      "loss": 3.8458,
+      "step": 2694144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.880776722182076e-06,
+      "loss": 3.8336,
+      "step": 2694656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.872407153475288e-06,
+      "loss": 3.8473,
+      "step": 2695168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.864021205964768e-06,
+      "loss": 3.8321,
+      "step": 2695680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.855635258454247e-06,
+      "loss": 3.8301,
+      "step": 2696192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.847249310943728e-06,
+      "loss": 3.8474,
+      "step": 2696704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.838863363433207e-06,
+      "loss": 3.839,
+      "step": 2697216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.830477415922688e-06,
+      "loss": 3.8413,
+      "step": 2697728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.822091468412167e-06,
+      "loss": 3.83,
+      "step": 2698240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.813705520901647e-06,
+      "loss": 3.8392,
+      "step": 2698752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.805335952194858e-06,
+      "loss": 3.8269,
+      "step": 2699264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.79696638348807e-06,
+      "loss": 3.8481,
+      "step": 2699776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.78858043597755e-06,
+      "loss": 3.8413,
+      "step": 2700288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.780194488467029e-06,
+      "loss": 3.8513,
+      "step": 2700800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.77180854095651e-06,
+      "loss": 3.8456,
+      "step": 2701312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.763422593445989e-06,
+      "loss": 3.8362,
+      "step": 2701824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.755036645935469e-06,
+      "loss": 3.8331,
+      "step": 2702336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.746650698424949e-06,
+      "loss": 3.8454,
+      "step": 2702848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.738264750914429e-06,
+      "loss": 3.832,
+      "step": 2703360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.729878803403909e-06,
+      "loss": 3.8366,
+      "step": 2703872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.721492855893389e-06,
+      "loss": 3.8427,
+      "step": 2704384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.7131232871866e-06,
+      "loss": 3.8423,
+      "step": 2704896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.7047373396760795e-06,
+      "loss": 3.8257,
+      "step": 2705408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.6963513921655594e-06,
+      "loss": 3.8315,
+      "step": 2705920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.68796544465504e-06,
+      "loss": 3.8263,
+      "step": 2706432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.679579497144519e-06,
+      "loss": 3.8523,
+      "step": 2706944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.671193549634e-06,
+      "loss": 3.8443,
+      "step": 2707456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.662807602123479e-06,
+      "loss": 3.8363,
+      "step": 2707968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.654438033416691e-06,
+      "loss": 3.8429,
+      "step": 2708480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.646052085906171e-06,
+      "loss": 3.838,
+      "step": 2708992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.63766613839565e-06,
+      "loss": 3.852,
+      "step": 2709504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.629280190885131e-06,
+      "loss": 3.8316,
+      "step": 2710016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.6209106221783416e-06,
+      "loss": 3.8527,
+      "step": 2710528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.6125246746678215e-06,
+      "loss": 3.8392,
+      "step": 2711040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.6041387271573015e-06,
+      "loss": 3.8443,
+      "step": 2711552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.595752779646781e-06,
+      "loss": 3.8427,
+      "step": 2712064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.587366832136261e-06,
+      "loss": 3.8393,
+      "step": 2712576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.578980884625741e-06,
+      "loss": 3.8492,
+      "step": 2713088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.570611315918953e-06,
+      "loss": 3.8422,
+      "step": 2713600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.562225368408432e-06,
+      "loss": 3.8317,
+      "step": 2714112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.553839420897913e-06,
+      "loss": 3.8414,
+      "step": 2714624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.545453473387392e-06,
+      "loss": 3.8319,
+      "step": 2715136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.537067525876873e-06,
+      "loss": 3.849,
+      "step": 2715648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.528681578366352e-06,
+      "loss": 3.843,
+      "step": 2716160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.520295630855832e-06,
+      "loss": 3.8289,
+      "step": 2716672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.511909683345312e-06,
+      "loss": 3.8436,
+      "step": 2717184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.503523735834792e-06,
+      "loss": 3.8349,
+      "step": 2717696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.495137788324272e-06,
+      "loss": 3.8409,
+      "step": 2718208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.486751840813752e-06,
+      "loss": 3.831,
+      "step": 2718720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.478382272106963e-06,
+      "loss": 3.8401,
+      "step": 2719232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.469996324596443e-06,
+      "loss": 3.8482,
+      "step": 2719744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.461610377085922e-06,
+      "loss": 3.836,
+      "step": 2720256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.453224429575403e-06,
+      "loss": 3.8268,
+      "step": 2720768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.444838482064882e-06,
+      "loss": 3.8393,
+      "step": 2721280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.436452534554363e-06,
+      "loss": 3.8357,
+      "step": 2721792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.428082965847574e-06,
+      "loss": 3.8488,
+      "step": 2722304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.419697018337054e-06,
+      "loss": 3.839,
+      "step": 2722816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.411311070826534e-06,
+      "loss": 3.8339,
+      "step": 2723328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.402925123316013e-06,
+      "loss": 3.8353,
+      "step": 2723840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.3945555546092245e-06,
+      "loss": 3.8453,
+      "step": 2724352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.3861696070987045e-06,
+      "loss": 3.8251,
+      "step": 2724864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.377783659588185e-06,
+      "loss": 3.8375,
+      "step": 2725376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.369397712077664e-06,
+      "loss": 3.8385,
+      "step": 2725888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.361011764567145e-06,
+      "loss": 3.8436,
+      "step": 2726400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.352625817056624e-06,
+      "loss": 3.8171,
+      "step": 2726912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.344239869546104e-06,
+      "loss": 3.8381,
+      "step": 2727424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.335853922035584e-06,
+      "loss": 3.8343,
+      "step": 2727936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.327467974525064e-06,
+      "loss": 3.8327,
+      "step": 2728448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.319098405818276e-06,
+      "loss": 3.8382,
+      "step": 2728960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.310712458307755e-06,
+      "loss": 3.8285,
+      "step": 2729472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.302326510797236e-06,
+      "loss": 3.8492,
+      "step": 2729984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.293940563286715e-06,
+      "loss": 3.8406,
+      "step": 2730496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.2855709945799265e-06,
+      "loss": 3.8318,
+      "step": 2731008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.2771850470694064e-06,
+      "loss": 3.8383,
+      "step": 2731520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.268799099558886e-06,
+      "loss": 3.8355,
+      "step": 2732032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.260413152048366e-06,
+      "loss": 3.8409,
+      "step": 2732544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.252027204537846e-06,
+      "loss": 3.8218,
+      "step": 2733056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.243641257027326e-06,
+      "loss": 3.847,
+      "step": 2733568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.235255309516806e-06,
+      "loss": 3.8327,
+      "step": 2734080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.226869362006285e-06,
+      "loss": 3.8389,
+      "step": 2734592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.218499793299497e-06,
+      "loss": 3.8342,
+      "step": 2735104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.210113845788977e-06,
+      "loss": 3.8441,
+      "step": 2735616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.201727898278457e-06,
+      "loss": 3.8342,
+      "step": 2736128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.1933583295716685e-06,
+      "loss": 3.8304,
+      "step": 2736640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.1849723820611485e-06,
+      "loss": 3.8251,
+      "step": 2737152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.176586434550628e-06,
+      "loss": 3.8496,
+      "step": 2737664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.1682004870401075e-06,
+      "loss": 3.8444,
+      "step": 2738176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.159814539529588e-06,
+      "loss": 3.8431,
+      "step": 2738688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.1514285920190674e-06,
+      "loss": 3.8318,
+      "step": 2739200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.143042644508548e-06,
+      "loss": 3.8456,
+      "step": 2739712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.134656696998027e-06,
+      "loss": 3.8439,
+      "step": 2740224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.126270749487507e-06,
+      "loss": 3.8349,
+      "step": 2740736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.117901180780719e-06,
+      "loss": 3.8347,
+      "step": 2741248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.109515233270198e-06,
+      "loss": 3.8442,
+      "step": 2741760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.101129285759679e-06,
+      "loss": 3.8433,
+      "step": 2742272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.092743338249158e-06,
+      "loss": 3.8473,
+      "step": 2742784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.084357390738639e-06,
+      "loss": 3.8306,
+      "step": 2743296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.0759878220318496e-06,
+      "loss": 3.8436,
+      "step": 2743808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.06760187452133e-06,
+      "loss": 3.8374,
+      "step": 2744320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.059232305814541e-06,
+      "loss": 3.8364,
+      "step": 2744832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.050846358304021e-06,
+      "loss": 3.8475,
+      "step": 2745344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.042460410793501e-06,
+      "loss": 3.8382,
+      "step": 2745856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.03407446328298e-06,
+      "loss": 3.839,
+      "step": 2746368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.025688515772461e-06,
+      "loss": 3.837,
+      "step": 2746880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.01730256826194e-06,
+      "loss": 3.8414,
+      "step": 2747392
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.026565074920654,
+      "eval_runtime": 296.8627,
+      "eval_samples_per_second": 1285.412,
+      "eval_steps_per_second": 40.17,
+      "step": 2747520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.008916620751421e-06,
+      "loss": 3.8316,
+      "step": 2747904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.0005306732409e-06,
+      "loss": 3.8412,
+      "step": 2748416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99214472573038e-06,
+      "loss": 3.8439,
+      "step": 2748928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.98375877821986e-06,
+      "loss": 3.8405,
+      "step": 2749440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.97537283070934e-06,
+      "loss": 3.8504,
+      "step": 2749952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.96698688319882e-06,
+      "loss": 3.8319,
+      "step": 2750464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9586009356883e-06,
+      "loss": 3.8377,
+      "step": 2750976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.95021498817778e-06,
+      "loss": 3.8296,
+      "step": 2751488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.94182904066726e-06,
+      "loss": 3.8473,
+      "step": 2752000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.93344309315674e-06,
+      "loss": 3.8409,
+      "step": 2752512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.92505714564622e-06,
+      "loss": 3.8431,
+      "step": 2753024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9166711981356995e-06,
+      "loss": 3.8405,
+      "step": 2753536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.908301629428911e-06,
+      "loss": 3.8393,
+      "step": 2754048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.89991568191839e-06,
+      "loss": 3.8304,
+      "step": 2754560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.89152973440787e-06,
+      "loss": 3.8335,
+      "step": 2755072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.88314378689735e-06,
+      "loss": 3.8247,
+      "step": 2755584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.87475783938683e-06,
+      "loss": 3.839,
+      "step": 2756096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.86637189187631e-06,
+      "loss": 3.8315,
+      "step": 2756608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.85798594436579e-06,
+      "loss": 3.8326,
+      "step": 2757120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.84959999685527e-06,
+      "loss": 3.8619,
+      "step": 2757632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.841230428148482e-06,
+      "loss": 3.8345,
+      "step": 2758144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.832860859441693e-06,
+      "loss": 3.8403,
+      "step": 2758656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.824474911931172e-06,
+      "loss": 3.8469,
+      "step": 2759168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.816088964420652e-06,
+      "loss": 3.8389,
+      "step": 2759680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.807703016910132e-06,
+      "loss": 3.8347,
+      "step": 2760192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.799317069399612e-06,
+      "loss": 3.8269,
+      "step": 2760704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.790931121889092e-06,
+      "loss": 3.8437,
+      "step": 2761216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.782545174378572e-06,
+      "loss": 3.8222,
+      "step": 2761728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.774159226868051e-06,
+      "loss": 3.8342,
+      "step": 2762240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.765773279357532e-06,
+      "loss": 3.8294,
+      "step": 2762752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.757403710650743e-06,
+      "loss": 3.8333,
+      "step": 2763264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.749017763140224e-06,
+      "loss": 3.8451,
+      "step": 2763776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.740631815629703e-06,
+      "loss": 3.8408,
+      "step": 2764288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.732245868119184e-06,
+      "loss": 3.8354,
+      "step": 2764800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.723876299412394e-06,
+      "loss": 3.8427,
+      "step": 2765312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.715490351901874e-06,
+      "loss": 3.8444,
+      "step": 2765824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.707104404391354e-06,
+      "loss": 3.8298,
+      "step": 2766336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.698718456880833e-06,
+      "loss": 3.8401,
+      "step": 2766848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.690332509370314e-06,
+      "loss": 3.8269,
+      "step": 2767360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.681979319467257e-06,
+      "loss": 3.8417,
+      "step": 2767872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.673593371956737e-06,
+      "loss": 3.8247,
+      "step": 2768384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.665223803249948e-06,
+      "loss": 3.8534,
+      "step": 2768896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.656837855739428e-06,
+      "loss": 3.8414,
+      "step": 2769408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.648451908228907e-06,
+      "loss": 3.8381,
+      "step": 2769920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.640065960718388e-06,
+      "loss": 3.8434,
+      "step": 2770432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.631680013207867e-06,
+      "loss": 3.8326,
+      "step": 2770944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.623294065697347e-06,
+      "loss": 3.8435,
+      "step": 2771456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.614908118186827e-06,
+      "loss": 3.8303,
+      "step": 2771968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.606522170676307e-06,
+      "loss": 3.8263,
+      "step": 2772480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.598136223165787e-06,
+      "loss": 3.8472,
+      "step": 2772992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.589750275655267e-06,
+      "loss": 3.8361,
+      "step": 2773504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.581364328144747e-06,
+      "loss": 3.8395,
+      "step": 2774016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.572978380634227e-06,
+      "loss": 3.8303,
+      "step": 2774528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.564608811927438e-06,
+      "loss": 3.8358,
+      "step": 2775040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.5562228644169185e-06,
+      "loss": 3.828,
+      "step": 2775552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.547836916906398e-06,
+      "loss": 3.8427,
+      "step": 2776064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.5394509693958785e-06,
+      "loss": 3.839,
+      "step": 2776576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.5310650218853576e-06,
+      "loss": 3.8516,
+      "step": 2777088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.5226790743748375e-06,
+      "loss": 3.8408,
+      "step": 2777600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.514309505668049e-06,
+      "loss": 3.8323,
+      "step": 2778112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.505923558157528e-06,
+      "loss": 3.8335,
+      "step": 2778624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.497537610647009e-06,
+      "loss": 3.8404,
+      "step": 2779136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.489151663136488e-06,
+      "loss": 3.8281,
+      "step": 2779648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.480765715625969e-06,
+      "loss": 3.8359,
+      "step": 2780160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.472379768115448e-06,
+      "loss": 3.8378,
+      "step": 2780672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.463993820604928e-06,
+      "loss": 3.8444,
+      "step": 2781184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.455607873094408e-06,
+      "loss": 3.8226,
+      "step": 2781696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.44723830438762e-06,
+      "loss": 3.829,
+      "step": 2782208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4388523568771e-06,
+      "loss": 3.8242,
+      "step": 2782720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.43048278817031e-06,
+      "loss": 3.8463,
+      "step": 2783232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.422096840659791e-06,
+      "loss": 3.8404,
+      "step": 2783744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.41371089314927e-06,
+      "loss": 3.8378,
+      "step": 2784256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.405324945638751e-06,
+      "loss": 3.8407,
+      "step": 2784768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.39693899812823e-06,
+      "loss": 3.8342,
+      "step": 2785280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.38855305061771e-06,
+      "loss": 3.8548,
+      "step": 2785792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.380183481910922e-06,
+      "loss": 3.8287,
+      "step": 2786304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.371797534400402e-06,
+      "loss": 3.8466,
+      "step": 2786816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.363411586889882e-06,
+      "loss": 3.8413,
+      "step": 2787328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.355025639379362e-06,
+      "loss": 3.8393,
+      "step": 2787840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.346639691868842e-06,
+      "loss": 3.8418,
+      "step": 2788352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.338270123162052e-06,
+      "loss": 3.8394,
+      "step": 2788864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.329884175651533e-06,
+      "loss": 3.8431,
+      "step": 2789376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.321498228141012e-06,
+      "loss": 3.8404,
+      "step": 2789888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.313112280630492e-06,
+      "loss": 3.8249,
+      "step": 2790400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.304726333119972e-06,
+      "loss": 3.8411,
+      "step": 2790912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.296340385609452e-06,
+      "loss": 3.8319,
+      "step": 2791424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.287954438098932e-06,
+      "loss": 3.8453,
+      "step": 2791936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.279568490588412e-06,
+      "loss": 3.8398,
+      "step": 2792448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.271198921881624e-06,
+      "loss": 3.8261,
+      "step": 2792960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.262812974371104e-06,
+      "loss": 3.8397,
+      "step": 2793472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.254443405664315e-06,
+      "loss": 3.8329,
+      "step": 2793984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2460574581537944e-06,
+      "loss": 3.8395,
+      "step": 2794496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.237671510643274e-06,
+      "loss": 3.8233,
+      "step": 2795008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.229285563132754e-06,
+      "loss": 3.84,
+      "step": 2795520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.220899615622234e-06,
+      "loss": 3.8519,
+      "step": 2796032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.212513668111714e-06,
+      "loss": 3.8315,
+      "step": 2796544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.204127720601194e-06,
+      "loss": 3.8199,
+      "step": 2797056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.195741773090673e-06,
+      "loss": 3.8363,
+      "step": 2797568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.187372204383885e-06,
+      "loss": 3.8362,
+      "step": 2798080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.178986256873365e-06,
+      "loss": 3.8464,
+      "step": 2798592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.170600309362845e-06,
+      "loss": 3.8399,
+      "step": 2799104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.162214361852325e-06,
+      "loss": 3.8312,
+      "step": 2799616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.153828414341805e-06,
+      "loss": 3.8317,
+      "step": 2800128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.145442466831285e-06,
+      "loss": 3.843,
+      "step": 2800640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.137072898124496e-06,
+      "loss": 3.8224,
+      "step": 2801152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.128686950613976e-06,
+      "loss": 3.8324,
+      "step": 2801664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1203010031034554e-06,
+      "loss": 3.8357,
+      "step": 2802176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.111915055592936e-06,
+      "loss": 3.8411,
+      "step": 2802688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.103529108082415e-06,
+      "loss": 3.8168,
+      "step": 2803200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.095159539375627e-06,
+      "loss": 3.8367,
+      "step": 2803712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.086773591865107e-06,
+      "loss": 3.8286,
+      "step": 2804224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.078387644354587e-06,
+      "loss": 3.8341,
+      "step": 2804736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.070001696844067e-06,
+      "loss": 3.8332,
+      "step": 2805248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.061615749333547e-06,
+      "loss": 3.8265,
+      "step": 2805760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.053229801823027e-06,
+      "loss": 3.8401,
+      "step": 2806272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.044843854312507e-06,
+      "loss": 3.8429,
+      "step": 2806784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.036474285605718e-06,
+      "loss": 3.8289,
+      "step": 2807296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0280883380951975e-06,
+      "loss": 3.8391,
+      "step": 2807808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0197023905846774e-06,
+      "loss": 3.8315,
+      "step": 2808320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.011316443074157e-06,
+      "loss": 3.842,
+      "step": 2808832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.002930495563637e-06,
+      "loss": 3.8222,
+      "step": 2809344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.994544548053117e-06,
+      "loss": 3.845,
+      "step": 2809856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.986174979346328e-06,
+      "loss": 3.8254,
+      "step": 2810368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.977789031835809e-06,
+      "loss": 3.8392,
+      "step": 2810880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.969403084325289e-06,
+      "loss": 3.8285,
+      "step": 2811392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.961017136814768e-06,
+      "loss": 3.8422,
+      "step": 2811904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.952631189304249e-06,
+      "loss": 3.8338,
+      "step": 2812416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9442616205974595e-06,
+      "loss": 3.8295,
+      "step": 2812928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9358756730869395e-06,
+      "loss": 3.8197,
+      "step": 2813440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9274897255764195e-06,
+      "loss": 3.8517,
+      "step": 2813952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.919103778065899e-06,
+      "loss": 3.8364,
+      "step": 2814464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.910717830555379e-06,
+      "loss": 3.841,
+      "step": 2814976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.902348261848591e-06,
+      "loss": 3.8303,
+      "step": 2815488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.89396231433807e-06,
+      "loss": 3.8449,
+      "step": 2816000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.88557636682755e-06,
+      "loss": 3.8458,
+      "step": 2816512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.87719041931703e-06,
+      "loss": 3.8297,
+      "step": 2817024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.868820850610242e-06,
+      "loss": 3.8327,
+      "step": 2817536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.860434903099722e-06,
+      "loss": 3.8402,
+      "step": 2818048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8520489555892016e-06,
+      "loss": 3.8432,
+      "step": 2818560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8436630080786815e-06,
+      "loss": 3.8438,
+      "step": 2819072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8352770605681615e-06,
+      "loss": 3.8278,
+      "step": 2819584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.826891113057641e-06,
+      "loss": 3.8404,
+      "step": 2820096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.818505165547121e-06,
+      "loss": 3.8336,
+      "step": 2820608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.810119218036601e-06,
+      "loss": 3.8365,
+      "step": 2821120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8017496493298126e-06,
+      "loss": 3.842,
+      "step": 2821632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.793363701819292e-06,
+      "loss": 3.8326,
+      "step": 2822144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7849777543087725e-06,
+      "loss": 3.8452,
+      "step": 2822656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.776591806798252e-06,
+      "loss": 3.8315,
+      "step": 2823168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7682058592877315e-06,
+      "loss": 3.8428,
+      "step": 2823680
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.025895118713379,
+      "eval_runtime": 296.803,
+      "eval_samples_per_second": 1285.671,
+      "eval_steps_per_second": 40.178,
+      "step": 2823840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.759819911777212e-06,
+      "loss": 3.8283,
+      "step": 2824192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7514339642666914e-06,
+      "loss": 3.8421,
+      "step": 2824704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.743048016756172e-06,
+      "loss": 3.8416,
+      "step": 2825216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7346620692456514e-06,
+      "loss": 3.8388,
+      "step": 2825728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.726276121735131e-06,
+      "loss": 3.8489,
+      "step": 2826240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7178901742246113e-06,
+      "loss": 3.8325,
+      "step": 2826752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.709504226714091e-06,
+      "loss": 3.8324,
+      "step": 2827264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.701118279203571e-06,
+      "loss": 3.8278,
+      "step": 2827776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.692732331693051e-06,
+      "loss": 3.8452,
+      "step": 2828288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6843463841825307e-06,
+      "loss": 3.8408,
+      "step": 2828800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.675960436672011e-06,
+      "loss": 3.8395,
+      "step": 2829312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6675744891614906e-06,
+      "loss": 3.8412,
+      "step": 2829824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6592049204547022e-06,
+      "loss": 3.8346,
+      "step": 2830336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6508189729441818e-06,
+      "loss": 3.8291,
+      "step": 2830848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.642433025433662e-06,
+      "loss": 3.8326,
+      "step": 2831360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6340470779231417e-06,
+      "loss": 3.823,
+      "step": 2831872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.625661130412621e-06,
+      "loss": 3.8326,
+      "step": 2832384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6172751829021016e-06,
+      "loss": 3.8272,
+      "step": 2832896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.608889235391581e-06,
+      "loss": 3.83,
+      "step": 2833408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6005032878810615e-06,
+      "loss": 3.8646,
+      "step": 2833920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5921337191742727e-06,
+      "loss": 3.8296,
+      "step": 2834432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5837477716637527e-06,
+      "loss": 3.8391,
+      "step": 2834944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5753618241532326e-06,
+      "loss": 3.8429,
+      "step": 2835456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.566975876642712e-06,
+      "loss": 3.8378,
+      "step": 2835968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5586063079359238e-06,
+      "loss": 3.8288,
+      "step": 2836480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5502203604254033e-06,
+      "loss": 3.8297,
+      "step": 2836992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5418344129148837e-06,
+      "loss": 3.8394,
+      "step": 2837504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5334484654043632e-06,
+      "loss": 3.8227,
+      "step": 2838016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5250625178938436e-06,
+      "loss": 3.8319,
+      "step": 2838528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5166929491870544e-06,
+      "loss": 3.8277,
+      "step": 2839040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5083070016765343e-06,
+      "loss": 3.8347,
+      "step": 2839552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4999210541660143e-06,
+      "loss": 3.8389,
+      "step": 2840064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4915351066554943e-06,
+      "loss": 3.8392,
+      "step": 2840576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4831491591449742e-06,
+      "loss": 3.8335,
+      "step": 2841088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4747795904381854e-06,
+      "loss": 3.8397,
+      "step": 2841600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.466393642927666e-06,
+      "loss": 3.8418,
+      "step": 2842112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4580240742208766e-06,
+      "loss": 3.8288,
+      "step": 2842624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.449638126710357e-06,
+      "loss": 3.8418,
+      "step": 2843136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4412521791998365e-06,
+      "loss": 3.8226,
+      "step": 2843648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.432866231689316e-06,
+      "loss": 3.8419,
+      "step": 2844160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4244802841787964e-06,
+      "loss": 3.822,
+      "step": 2844672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.416094336668276e-06,
+      "loss": 3.8525,
+      "step": 2845184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4077083891577563e-06,
+      "loss": 3.8405,
+      "step": 2845696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.399322441647236e-06,
+      "loss": 3.8369,
+      "step": 2846208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.390936494136716e-06,
+      "loss": 3.8413,
+      "step": 2846720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.382550546626196e-06,
+      "loss": 3.8342,
+      "step": 2847232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.374180977919407e-06,
+      "loss": 3.8376,
+      "step": 2847744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3657950304088874e-06,
+      "loss": 3.8259,
+      "step": 2848256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.357409082898367e-06,
+      "loss": 3.8187,
+      "step": 2848768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3490231353878473e-06,
+      "loss": 3.8509,
+      "step": 2849280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.340653566681058e-06,
+      "loss": 3.8346,
+      "step": 2849792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3322676191705384e-06,
+      "loss": 3.843,
+      "step": 2850304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3238980504637492e-06,
+      "loss": 3.8267,
+      "step": 2850816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3155121029532296e-06,
+      "loss": 3.8342,
+      "step": 2851328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3071261554427096e-06,
+      "loss": 3.8251,
+      "step": 2851840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.298740207932189e-06,
+      "loss": 3.8396,
+      "step": 2852352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2903542604216695e-06,
+      "loss": 3.8327,
+      "step": 2852864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.281968312911149e-06,
+      "loss": 3.8473,
+      "step": 2853376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2735823654006294e-06,
+      "loss": 3.8445,
+      "step": 2853888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.265196417890109e-06,
+      "loss": 3.8257,
+      "step": 2854400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2568268491833206e-06,
+      "loss": 3.8309,
+      "step": 2854912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2484409016728e-06,
+      "loss": 3.842,
+      "step": 2855424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2400549541622796e-06,
+      "loss": 3.8242,
+      "step": 2855936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.23166900665176e-06,
+      "loss": 3.8324,
+      "step": 2856448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2232994379449712e-06,
+      "loss": 3.8379,
+      "step": 2856960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.214913490434451e-06,
+      "loss": 3.8398,
+      "step": 2857472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.206527542923931e-06,
+      "loss": 3.8227,
+      "step": 2857984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.198141595413411e-06,
+      "loss": 3.8273,
+      "step": 2858496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.189755647902891e-06,
+      "loss": 3.8218,
+      "step": 2859008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1813697003923706e-06,
+      "loss": 3.8429,
+      "step": 2859520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.172983752881851e-06,
+      "loss": 3.8366,
+      "step": 2860032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1645978053713305e-06,
+      "loss": 3.8355,
+      "step": 2860544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.156228236664542e-06,
+      "loss": 3.8409,
+      "step": 2861056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1478422891540217e-06,
+      "loss": 3.8305,
+      "step": 2861568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1394727204472333e-06,
+      "loss": 3.8526,
+      "step": 2862080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1311031517404445e-06,
+      "loss": 3.8237,
+      "step": 2862592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1227172042299245e-06,
+      "loss": 3.843,
+      "step": 2863104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1143312567194044e-06,
+      "loss": 3.8383,
+      "step": 2863616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1059453092088844e-06,
+      "loss": 3.8362,
+      "step": 2864128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0975593616983643e-06,
+      "loss": 3.8415,
+      "step": 2864640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0891734141878443e-06,
+      "loss": 3.8361,
+      "step": 2865152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0807874666773242e-06,
+      "loss": 3.8383,
+      "step": 2865664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0724015191668038e-06,
+      "loss": 3.8419,
+      "step": 2866176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0640155716562837e-06,
+      "loss": 3.8217,
+      "step": 2866688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0556296241457637e-06,
+      "loss": 3.8426,
+      "step": 2867200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.047260055438975e-06,
+      "loss": 3.8286,
+      "step": 2867712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.038874107928455e-06,
+      "loss": 3.8437,
+      "step": 2868224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.030488160417935e-06,
+      "loss": 3.835,
+      "step": 2868736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0221022129074143e-06,
+      "loss": 3.8252,
+      "step": 2869248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0137162653968943e-06,
+      "loss": 3.8358,
+      "step": 2869760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.005346696690106e-06,
+      "loss": 3.8369,
+      "step": 2870272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.996960749179586e-06,
+      "loss": 3.8355,
+      "step": 2870784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.988574801669066e-06,
+      "loss": 3.8223,
+      "step": 2871296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.980205232962277e-06,
+      "loss": 3.8397,
+      "step": 2871808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.971819285451757e-06,
+      "loss": 3.8452,
+      "step": 2872320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.963433337941237e-06,
+      "loss": 3.8338,
+      "step": 2872832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.955047390430717e-06,
+      "loss": 3.816,
+      "step": 2873344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9466614429201965e-06,
+      "loss": 3.8341,
+      "step": 2873856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9382754954096764e-06,
+      "loss": 3.8336,
+      "step": 2874368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9298895478991564e-06,
+      "loss": 3.8405,
+      "step": 2874880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9215036003886363e-06,
+      "loss": 3.8406,
+      "step": 2875392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9131176528781163e-06,
+      "loss": 3.8314,
+      "step": 2875904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9047317053675962e-06,
+      "loss": 3.8273,
+      "step": 2876416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.896345757857076e-06,
+      "loss": 3.8406,
+      "step": 2876928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8879761891502874e-06,
+      "loss": 3.82,
+      "step": 2877440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8795902416397674e-06,
+      "loss": 3.8317,
+      "step": 2877952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8712206729329786e-06,
+      "loss": 3.8306,
+      "step": 2878464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8628347254224585e-06,
+      "loss": 3.8399,
+      "step": 2878976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8544487779119385e-06,
+      "loss": 3.8191,
+      "step": 2879488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8460628304014184e-06,
+      "loss": 3.8301,
+      "step": 2880000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8376768828908984e-06,
+      "loss": 3.8316,
+      "step": 2880512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.829290935380378e-06,
+      "loss": 3.8295,
+      "step": 2881024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.820904987869858e-06,
+      "loss": 3.8315,
+      "step": 2881536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.812519040359338e-06,
+      "loss": 3.8223,
+      "step": 2882048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8041494716525495e-06,
+      "loss": 3.8398,
+      "step": 2882560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7957635241420294e-06,
+      "loss": 3.8407,
+      "step": 2883072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7873939554352407e-06,
+      "loss": 3.8262,
+      "step": 2883584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7790080079247206e-06,
+      "loss": 3.8396,
+      "step": 2884096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7706220604142006e-06,
+      "loss": 3.8281,
+      "step": 2884608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.76223611290368e-06,
+      "loss": 3.8407,
+      "step": 2885120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.75385016539316e-06,
+      "loss": 3.825,
+      "step": 2885632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.74546421788264e-06,
+      "loss": 3.8388,
+      "step": 2886144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.73707827037212e-06,
+      "loss": 3.8232,
+      "step": 2886656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.728708701665331e-06,
+      "loss": 3.8368,
+      "step": 2887168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.720322754154811e-06,
+      "loss": 3.8272,
+      "step": 2887680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.711936806644291e-06,
+      "loss": 3.8398,
+      "step": 2888192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.703550859133771e-06,
+      "loss": 3.8302,
+      "step": 2888704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.695164911623251e-06,
+      "loss": 3.8335,
+      "step": 2889216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6867953429164622e-06,
+      "loss": 3.8149,
+      "step": 2889728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.678409395405942e-06,
+      "loss": 3.8496,
+      "step": 2890240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.670023447895422e-06,
+      "loss": 3.8314,
+      "step": 2890752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.661637500384902e-06,
+      "loss": 3.839,
+      "step": 2891264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6532679316781133e-06,
+      "loss": 3.8265,
+      "step": 2891776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6448819841675933e-06,
+      "loss": 3.8433,
+      "step": 2892288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.636496036657073e-06,
+      "loss": 3.8408,
+      "step": 2892800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6281100891465527e-06,
+      "loss": 3.8299,
+      "step": 2893312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.619740520439765e-06,
+      "loss": 3.8347,
+      "step": 2893824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6113545729292443e-06,
+      "loss": 3.8374,
+      "step": 2894336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6029686254187243e-06,
+      "loss": 3.8395,
+      "step": 2894848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5945826779082042e-06,
+      "loss": 3.8505,
+      "step": 2895360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.586196730397684e-06,
+      "loss": 3.8186,
+      "step": 2895872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5778271616908954e-06,
+      "loss": 3.843,
+      "step": 2896384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5694412141803754e-06,
+      "loss": 3.8305,
+      "step": 2896896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5610552666698553e-06,
+      "loss": 3.8384,
+      "step": 2897408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.552669319159335e-06,
+      "loss": 3.8374,
+      "step": 2897920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5442997504525465e-06,
+      "loss": 3.8264,
+      "step": 2898432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5359138029420265e-06,
+      "loss": 3.8459,
+      "step": 2898944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5275278554315064e-06,
+      "loss": 3.8318,
+      "step": 2899456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5191419079209864e-06,
+      "loss": 3.8395,
+      "step": 2899968
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.025328159332275,
+      "eval_runtime": 295.0626,
+      "eval_samples_per_second": 1293.254,
+      "eval_steps_per_second": 40.415,
+      "step": 2900160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.5107559604104663e-06,
+      "loss": 3.8309,
+      "step": 2900480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.5023700128999463e-06,
+      "loss": 3.8356,
+      "step": 2900992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4940004441931575e-06,
+      "loss": 3.8425,
+      "step": 2901504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4856144966826374e-06,
+      "loss": 3.8377,
+      "step": 2902016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.477228549172117e-06,
+      "loss": 3.843,
+      "step": 2902528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.468842601661597e-06,
+      "loss": 3.8362,
+      "step": 2903040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.460456654151077e-06,
+      "loss": 3.8254,
+      "step": 2903552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.452070706640557e-06,
+      "loss": 3.8293,
+      "step": 2904064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.443701137933768e-06,
+      "loss": 3.8383,
+      "step": 2904576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.435315190423248e-06,
+      "loss": 3.8424,
+      "step": 2905088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.426929242912728e-06,
+      "loss": 3.8343,
+      "step": 2905600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.418543295402208e-06,
+      "loss": 3.8414,
+      "step": 2906112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.410173726695419e-06,
+      "loss": 3.8268,
+      "step": 2906624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.401787779184899e-06,
+      "loss": 3.8312,
+      "step": 2907136
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.393401831674379e-06,
+      "loss": 3.8322,
+      "step": 2907648
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.385015884163859e-06,
+      "loss": 3.8222,
+      "step": 2908160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.376629936653339e-06,
+      "loss": 3.8355,
+      "step": 2908672
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3682439891428185e-06,
+      "loss": 3.8235,
+      "step": 2909184
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3598580416322984e-06,
+      "loss": 3.8277,
+      "step": 2909696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3514720941217784e-06,
+      "loss": 3.8643,
+      "step": 2910208
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3431025254149896e-06,
+      "loss": 3.8288,
+      "step": 2910720
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3347165779044696e-06,
+      "loss": 3.8338,
+      "step": 2911232
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3263306303939495e-06,
+      "loss": 3.8401,
+      "step": 2911744
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3179446828834295e-06,
+      "loss": 3.845,
+      "step": 2912256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3095587353729094e-06,
+      "loss": 3.8204,
+      "step": 2912768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3011727878623894e-06,
+      "loss": 3.8302,
+      "step": 2913280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2927868403518693e-06,
+      "loss": 3.835,
+      "step": 2913792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2844008928413493e-06,
+      "loss": 3.8259,
+      "step": 2914304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2760313241345605e-06,
+      "loss": 3.8289,
+      "step": 2914816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2676453766240405e-06,
+      "loss": 3.8237,
+      "step": 2915328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2592594291135204e-06,
+      "loss": 3.8343,
+      "step": 2915840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.250873481603e-06,
+      "loss": 3.8391,
+      "step": 2916352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2425039128962116e-06,
+      "loss": 3.8348,
+      "step": 2916864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.234117965385691e-06,
+      "loss": 3.8316,
+      "step": 2917376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.225748396678903e-06,
+      "loss": 3.8347,
+      "step": 2917888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2173624491683827e-06,
+      "loss": 3.8383,
+      "step": 2918400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2089765016578627e-06,
+      "loss": 3.8231,
+      "step": 2918912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2005905541473426e-06,
+      "loss": 3.8434,
+      "step": 2919424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1922046066368226e-06,
+      "loss": 3.8186,
+      "step": 2919936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1838186591263025e-06,
+      "loss": 3.8384,
+      "step": 2920448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.175432711615782e-06,
+      "loss": 3.8193,
+      "step": 2920960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.167046764105262e-06,
+      "loss": 3.8492,
+      "step": 2921472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1586771953984732e-06,
+      "loss": 3.8337,
+      "step": 2921984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.150291247887953e-06,
+      "loss": 3.8364,
+      "step": 2922496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.141905300377433e-06,
+      "loss": 3.8377,
+      "step": 2923008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.133535731670645e-06,
+      "loss": 3.8333,
+      "step": 2923520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1251497841601248e-06,
+      "loss": 3.8376,
+      "step": 2924032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1167638366496047e-06,
+      "loss": 3.8269,
+      "step": 2924544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1083778891390842e-06,
+      "loss": 3.8144,
+      "step": 2925056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.099991941628564e-06,
+      "loss": 3.8513,
+      "step": 2925568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0916223729217754e-06,
+      "loss": 3.8309,
+      "step": 2926080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0832364254112554e-06,
+      "loss": 3.8451,
+      "step": 2926592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0748504779007353e-06,
+      "loss": 3.8232,
+      "step": 2927104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0664645303902153e-06,
+      "loss": 3.8315,
+      "step": 2927616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0580949616834265e-06,
+      "loss": 3.8213,
+      "step": 2928128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0497090141729064e-06,
+      "loss": 3.8411,
+      "step": 2928640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0413230666623864e-06,
+      "loss": 3.8359,
+      "step": 2929152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0329371191518664e-06,
+      "loss": 3.8442,
+      "step": 2929664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0245511716413463e-06,
+      "loss": 3.8411,
+      "step": 2930176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0161652241308263e-06,
+      "loss": 3.8245,
+      "step": 2930688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0077792766203062e-06,
+      "loss": 3.8262,
+      "step": 2931200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.999393329109786e-06,
+      "loss": 3.8441,
+      "step": 2931712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9910237604029974e-06,
+      "loss": 3.8229,
+      "step": 2932224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9826378128924774e-06,
+      "loss": 3.8285,
+      "step": 2932736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.974251865381957e-06,
+      "loss": 3.8349,
+      "step": 2933248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9658822966751685e-06,
+      "loss": 3.841,
+      "step": 2933760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.957496349164648e-06,
+      "loss": 3.821,
+      "step": 2934272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.949110401654128e-06,
+      "loss": 3.8243,
+      "step": 2934784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.940724454143608e-06,
+      "loss": 3.8228,
+      "step": 2935296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.932338506633088e-06,
+      "loss": 3.8328,
+      "step": 2935808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.923952559122568e-06,
+      "loss": 3.8388,
+      "step": 2936320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9155829904157795e-06,
+      "loss": 3.8344,
+      "step": 2936832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9071970429052593e-06,
+      "loss": 3.8359,
+      "step": 2937344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.898811095394739e-06,
+      "loss": 3.8268,
+      "step": 2937856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.890425147884219e-06,
+      "loss": 3.8528,
+      "step": 2938368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.882039200373699e-06,
+      "loss": 3.8219,
+      "step": 2938880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8736532528631789e-06,
+      "loss": 3.8416,
+      "step": 2939392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8652673053526584e-06,
+      "loss": 3.8363,
+      "step": 2939904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8568977366458702e-06,
+      "loss": 3.8346,
+      "step": 2940416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8485281679390815e-06,
+      "loss": 3.8399,
+      "step": 2940928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8401422204285614e-06,
+      "loss": 3.8385,
+      "step": 2941440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.831756272918041e-06,
+      "loss": 3.8321,
+      "step": 2941952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.823370325407521e-06,
+      "loss": 3.8402,
+      "step": 2942464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.814984377897001e-06,
+      "loss": 3.8233,
+      "step": 2942976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.806598430386481e-06,
+      "loss": 3.8359,
+      "step": 2943488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.798212482875961e-06,
+      "loss": 3.8301,
+      "step": 2944000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7898265353654405e-06,
+      "loss": 3.8411,
+      "step": 2944512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7814569666586522e-06,
+      "loss": 3.8334,
+      "step": 2945024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.773071019148132e-06,
+      "loss": 3.8239,
+      "step": 2945536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7646850716376119e-06,
+      "loss": 3.831,
+      "step": 2946048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7562991241270918e-06,
+      "loss": 3.8336,
+      "step": 2946560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.747929555420303e-06,
+      "loss": 3.8329,
+      "step": 2947072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.739543607909783e-06,
+      "loss": 3.8217,
+      "step": 2947584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.731157660399263e-06,
+      "loss": 3.8385,
+      "step": 2948096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7227717128887429e-06,
+      "loss": 3.8403,
+      "step": 2948608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7143857653782226e-06,
+      "loss": 3.8349,
+      "step": 2949120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7060161966714343e-06,
+      "loss": 3.8133,
+      "step": 2949632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6976302491609138e-06,
+      "loss": 3.8318,
+      "step": 2950144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6892443016503938e-06,
+      "loss": 3.8285,
+      "step": 2950656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6808583541398737e-06,
+      "loss": 3.8406,
+      "step": 2951168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6724887854330851e-06,
+      "loss": 3.8406,
+      "step": 2951680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.664102837922565e-06,
+      "loss": 3.827,
+      "step": 2952192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.655716890412045e-06,
+      "loss": 3.8272,
+      "step": 2952704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.647330942901525e-06,
+      "loss": 3.839,
+      "step": 2953216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6389449953910045e-06,
+      "loss": 3.8189,
+      "step": 2953728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6305590478804845e-06,
+      "loss": 3.833,
+      "step": 2954240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.622189479173696e-06,
+      "loss": 3.8249,
+      "step": 2954752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6138035316631759e-06,
+      "loss": 3.8428,
+      "step": 2955264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.605433962956387e-06,
+      "loss": 3.8156,
+      "step": 2955776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.597048015445867e-06,
+      "loss": 3.8248,
+      "step": 2956288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.588662067935347e-06,
+      "loss": 3.827,
+      "step": 2956800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.580276120424827e-06,
+      "loss": 3.8349,
+      "step": 2957312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.571890172914307e-06,
+      "loss": 3.8255,
+      "step": 2957824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5635042254037867e-06,
+      "loss": 3.824,
+      "step": 2958336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5551182778932666e-06,
+      "loss": 3.8357,
+      "step": 2958848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5467323303827466e-06,
+      "loss": 3.8389,
+      "step": 2959360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5383627616759578e-06,
+      "loss": 3.8245,
+      "step": 2959872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5299768141654377e-06,
+      "loss": 3.8397,
+      "step": 2960384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5215908666549177e-06,
+      "loss": 3.8238,
+      "step": 2960896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5132049191443977e-06,
+      "loss": 3.8355,
+      "step": 2961408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5048189716338776e-06,
+      "loss": 3.827,
+      "step": 2961920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4964330241233574e-06,
+      "loss": 3.8346,
+      "step": 2962432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4880470766128373e-06,
+      "loss": 3.8224,
+      "step": 2962944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4796775079060485e-06,
+      "loss": 3.832,
+      "step": 2963456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4712915603955285e-06,
+      "loss": 3.8323,
+      "step": 2963968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4629056128850084e-06,
+      "loss": 3.8323,
+      "step": 2964480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4545196653744884e-06,
+      "loss": 3.8275,
+      "step": 2964992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4461337178639681e-06,
+      "loss": 3.8323,
+      "step": 2965504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4377641491571796e-06,
+      "loss": 3.8161,
+      "step": 2966016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4293782016466595e-06,
+      "loss": 3.8464,
+      "step": 2966528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4209922541361395e-06,
+      "loss": 3.831,
+      "step": 2967040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4126063066256194e-06,
+      "loss": 3.8367,
+      "step": 2967552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4042367379188306e-06,
+      "loss": 3.8268,
+      "step": 2968064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3958507904083106e-06,
+      "loss": 3.8455,
+      "step": 2968576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3874648428977903e-06,
+      "loss": 3.8344,
+      "step": 2969088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3790788953872703e-06,
+      "loss": 3.8281,
+      "step": 2969600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3707093266804817e-06,
+      "loss": 3.8311,
+      "step": 2970112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3623233791699617e-06,
+      "loss": 3.8346,
+      "step": 2970624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3539374316594414e-06,
+      "loss": 3.8362,
+      "step": 2971136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3455678629526528e-06,
+      "loss": 3.8451,
+      "step": 2971648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3371819154421328e-06,
+      "loss": 3.8178,
+      "step": 2972160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3287959679316128e-06,
+      "loss": 3.8395,
+      "step": 2972672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3204100204210925e-06,
+      "loss": 3.8293,
+      "step": 2973184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3120240729105725e-06,
+      "loss": 3.8389,
+      "step": 2973696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3036381254000524e-06,
+      "loss": 3.8329,
+      "step": 2974208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2952521778895322e-06,
+      "loss": 3.8283,
+      "step": 2974720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2868662303790121e-06,
+      "loss": 3.8406,
+      "step": 2975232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2784802828684919e-06,
+      "loss": 3.8333,
+      "step": 2975744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2701107141617035e-06,
+      "loss": 3.8371,
+      "step": 2976256
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.024765968322754,
+      "eval_runtime": 294.8903,
+      "eval_samples_per_second": 1294.01,
+      "eval_steps_per_second": 40.439,
+      "step": 2976480
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.2617247666511832e-06,
+      "loss": 3.8335,
+      "step": 2976768
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.2533388191406632e-06,
+      "loss": 3.8316,
+      "step": 2977280
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.2449528716301431e-06,
+      "loss": 3.8389,
+      "step": 2977792
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.2365669241196229e-06,
+      "loss": 3.8371,
+      "step": 2978304
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.2281809766091028e-06,
+      "loss": 3.8449,
+      "step": 2978816
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.2198114079023143e-06,
+      "loss": 3.8295,
+      "step": 2979328
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.2114254603917942e-06,
+      "loss": 3.824,
+      "step": 2979840
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.203039512881274e-06,
+      "loss": 3.8232,
+      "step": 2980352
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.194653565370754e-06,
+      "loss": 3.8388,
+      "step": 2980864
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1862676178602337e-06,
+      "loss": 3.8408,
+      "step": 2981376
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1778980491534453e-06,
+      "loss": 3.8287,
+      "step": 2981888
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.169512101642925e-06,
+      "loss": 3.8387,
+      "step": 2982400
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1611425329361365e-06,
+      "loss": 3.8302,
+      "step": 2982912
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1527565854256162e-06,
+      "loss": 3.8289,
+      "step": 2983424
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1443706379150962e-06,
+      "loss": 3.8281,
+      "step": 2983936
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1359846904045761e-06,
+      "loss": 3.8238,
+      "step": 2984448
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.127598742894056e-06,
+      "loss": 3.8296,
+      "step": 2984960
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.119212795383536e-06,
+      "loss": 3.8246,
+      "step": 2985472
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1108268478730158e-06,
+      "loss": 3.8246,
+      "step": 2985984
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1024409003624957e-06,
+      "loss": 3.8583,
+      "step": 2986496
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.0940549528519757e-06,
+      "loss": 3.833,
+      "step": 2987008
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.0856690053414554e-06,
+      "loss": 3.8303,
+      "step": 2987520
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.0772830578309354e-06,
+      "loss": 3.8352,
+      "step": 2988032
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.0689134891241468e-06,
+      "loss": 3.8387,
+      "step": 2988544
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.0605439204173583e-06,
+      "loss": 3.8209,
+      "step": 2989056
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.052157972906838e-06,
+      "loss": 3.8299,
+      "step": 2989568
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.043772025396318e-06,
+      "loss": 3.8306,
+      "step": 2990080
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.035386077885798e-06,
+      "loss": 3.8273,
+      "step": 2990592
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.0270001303752779e-06,
+      "loss": 3.8221,
+      "step": 2991104
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.0186141828647576e-06,
+      "loss": 3.8239,
+      "step": 2991616
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0102282353542376e-06,
+      "loss": 3.8317,
+      "step": 2992128
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0018422878437175e-06,
+      "loss": 3.8388,
+      "step": 2992640
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.934563403331973e-07,
+      "loss": 3.8336,
+      "step": 2993152
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.850867716264087e-07,
+      "loss": 3.8283,
+      "step": 2993664
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.767172029196201e-07,
+      "loss": 3.835,
+      "step": 2994176
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.683312554091e-07,
+      "loss": 3.835,
+      "step": 2994688
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.599453078985798e-07,
+      "loss": 3.8284,
+      "step": 2995200
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.515593603880598e-07,
+      "loss": 3.8354,
+      "step": 2995712
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.431734128775396e-07,
+      "loss": 3.8202,
+      "step": 2996224
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.347874653670196e-07,
+      "loss": 3.8347,
+      "step": 2996736
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.264015178564995e-07,
+      "loss": 3.8217,
+      "step": 2997248
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.180319491497108e-07,
+      "loss": 3.8439,
+      "step": 2997760
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.096460016391908e-07,
+      "loss": 3.8344,
+      "step": 2998272
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.012600541286705e-07,
+      "loss": 3.8334,
+      "step": 2998784
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.928741066181506e-07,
+      "loss": 3.8395,
+      "step": 2999296
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.844881591076304e-07,
+      "loss": 3.8278,
+      "step": 2999808
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.761022115971103e-07,
+      "loss": 3.8364,
+      "step": 3000320
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.677162640865902e-07,
+      "loss": 3.8282,
+      "step": 3000832
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.593303165760701e-07,
+      "loss": 3.8076,
+      "step": 3001344
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.509443690655501e-07,
+      "loss": 3.8501,
+      "step": 3001856
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.425748003587614e-07,
+      "loss": 3.8281,
+      "step": 3002368
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.341888528482413e-07,
+      "loss": 3.8433,
+      "step": 3002880
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.258029053377211e-07,
+      "loss": 3.8244,
+      "step": 3003392
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.17416957827201e-07,
+      "loss": 3.8297,
+      "step": 3003904
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.090310103166809e-07,
+      "loss": 3.8174,
+      "step": 3004416
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 8.006450628061608e-07,
+      "loss": 3.8393,
+      "step": 3004928
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.922754940993722e-07,
+      "loss": 3.8318,
+      "step": 3005440
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.839059253925836e-07,
+      "loss": 3.8389,
+      "step": 3005952
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.755199778820636e-07,
+      "loss": 3.8443,
+      "step": 3006464
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.671340303715434e-07,
+      "loss": 3.825,
+      "step": 3006976
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.587480828610233e-07,
+      "loss": 3.8239,
+      "step": 3007488
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.503621353505031e-07,
+      "loss": 3.8436,
+      "step": 3008000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.419761878399831e-07,
+      "loss": 3.8216,
+      "step": 3008512
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.33590240329463e-07,
+      "loss": 3.8252,
+      "step": 3009024
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.252042928189429e-07,
+      "loss": 3.8338,
+      "step": 3009536
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.168183453084227e-07,
+      "loss": 3.8378,
+      "step": 3010048
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.084323977979027e-07,
+      "loss": 3.8204,
+      "step": 3010560
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.00062829091114e-07,
+      "loss": 3.8253,
+      "step": 3011072
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.916768815805939e-07,
+      "loss": 3.8176,
+      "step": 3011584
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.832909340700738e-07,
+      "loss": 3.8306,
+      "step": 3012096
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.749049865595536e-07,
+      "loss": 3.8377,
+      "step": 3012608
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.665190390490336e-07,
+      "loss": 3.8346,
+      "step": 3013120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.581330915385136e-07,
+      "loss": 3.8325,
+      "step": 3013632
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.497799016354563e-07,
+      "loss": 3.8311,
+      "step": 3014144
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.413939541249362e-07,
+      "loss": 3.8504,
+      "step": 3014656
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.330080066144161e-07,
+      "loss": 3.8206,
+      "step": 3015168
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.24622059103896e-07,
+      "loss": 3.8418,
+      "step": 3015680
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.16236111593376e-07,
+      "loss": 3.8317,
+      "step": 3016192
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.078501640828558e-07,
+      "loss": 3.8272,
+      "step": 3016704
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.994642165723357e-07,
+      "loss": 3.8412,
+      "step": 3017216
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.910782690618156e-07,
+      "loss": 3.8353,
+      "step": 3017728
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.827087003550269e-07,
+      "loss": 3.8313,
+      "step": 3018240
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.743227528445069e-07,
+      "loss": 3.8365,
+      "step": 3018752
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.659368053339868e-07,
+      "loss": 3.8237,
+      "step": 3019264
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.575508578234667e-07,
+      "loss": 3.8331,
+      "step": 3019776
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.491649103129465e-07,
+      "loss": 3.8295,
+      "step": 3020288
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.407789628024265e-07,
+      "loss": 3.841,
+      "step": 3020800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.323930152919063e-07,
+      "loss": 3.8293,
+      "step": 3021312
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.240234465851178e-07,
+      "loss": 3.8263,
+      "step": 3021824
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 5.156374990745976e-07,
+      "loss": 3.8235,
+      "step": 3022336
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 5.072515515640775e-07,
+      "loss": 3.8359,
+      "step": 3022848
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.988656040535574e-07,
+      "loss": 3.8338,
+      "step": 3023360
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.904796565430373e-07,
+      "loss": 3.82,
+      "step": 3023872
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.820937090325172e-07,
+      "loss": 3.8377,
+      "step": 3024384
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.737241403257286e-07,
+      "loss": 3.8367,
+      "step": 3024896
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.6533819281520845e-07,
+      "loss": 3.8307,
+      "step": 3025408
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.5695224530468836e-07,
+      "loss": 3.8142,
+      "step": 3025920
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4856629779416826e-07,
+      "loss": 3.8302,
+      "step": 3026432
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4019672908737963e-07,
+      "loss": 3.8243,
+      "step": 3026944
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3181078157685953e-07,
+      "loss": 3.8412,
+      "step": 3027456
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.234248340663394e-07,
+      "loss": 3.8395,
+      "step": 3027968
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.150388865558193e-07,
+      "loss": 3.8264,
+      "step": 3028480
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.0665293904529924e-07,
+      "loss": 3.8279,
+      "step": 3028992
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.982669915347791e-07,
+      "loss": 3.8309,
+      "step": 3029504
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8989742282799046e-07,
+      "loss": 3.8234,
+      "step": 3030016
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.8151147531747037e-07,
+      "loss": 3.8284,
+      "step": 3030528
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.7314190661068174e-07,
+      "loss": 3.8226,
+      "step": 3031040
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.6475595910016164e-07,
+      "loss": 3.844,
+      "step": 3031552
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5637001158964155e-07,
+      "loss": 3.8144,
+      "step": 3032064
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.479840640791214e-07,
+      "loss": 3.8251,
+      "step": 3032576
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.3959811656860135e-07,
+      "loss": 3.8231,
+      "step": 3033088
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.3122854786181267e-07,
+      "loss": 3.8347,
+      "step": 3033600
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2284260035129263e-07,
+      "loss": 3.821,
+      "step": 3034112
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.144566528407725e-07,
+      "loss": 3.8222,
+      "step": 3034624
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.060707053302524e-07,
+      "loss": 3.8348,
+      "step": 3035136
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.976847578197323e-07,
+      "loss": 3.8384,
+      "step": 3035648
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.892988103092122e-07,
+      "loss": 3.8212,
+      "step": 3036160
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8091286279869203e-07,
+      "loss": 3.8354,
+      "step": 3036672
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.72526915288172e-07,
+      "loss": 3.8263,
+      "step": 3037184
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.6414096777765184e-07,
+      "loss": 3.8319,
+      "step": 3037696
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.5577139907086327e-07,
+      "loss": 3.8262,
+      "step": 3038208
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.473854515603431e-07,
+      "loss": 3.8289,
+      "step": 3038720
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.38999504049823e-07,
+      "loss": 3.8241,
+      "step": 3039232
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.3061355653930295e-07,
+      "loss": 3.8307,
+      "step": 3039744
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.2222760902878282e-07,
+      "loss": 3.8342,
+      "step": 3040256
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.138416615182627e-07,
+      "loss": 3.8283,
+      "step": 3040768
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.054720928114741e-07,
+      "loss": 3.8269,
+      "step": 3041280
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.9708614530095397e-07,
+      "loss": 3.8295,
+      "step": 3041792
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.8870019779043388e-07,
+      "loss": 3.8146,
+      "step": 3042304
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.8031425027991378e-07,
+      "loss": 3.846,
+      "step": 3042816
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.7192830276939368e-07,
+      "loss": 3.8262,
+      "step": 3043328
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.6354235525887356e-07,
+      "loss": 3.8358,
+      "step": 3043840
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.5517278655208493e-07,
+      "loss": 3.823,
+      "step": 3044352
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.4678683904156483e-07,
+      "loss": 3.8419,
+      "step": 3044864
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.384008915310447e-07,
+      "loss": 3.8341,
+      "step": 3045376
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.300149440205246e-07,
+      "loss": 3.8274,
+      "step": 3045888
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.2162899651000452e-07,
+      "loss": 3.831,
+      "step": 3046400
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.132430489994844e-07,
+      "loss": 3.8301,
+      "step": 3046912
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0487348029269578e-07,
+      "loss": 3.8336,
+      "step": 3047424
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 9.648753278217567e-08,
+      "loss": 3.8488,
+      "step": 3047936
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 8.810158527165557e-08,
+      "loss": 3.813,
+      "step": 3048448
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 7.971563776113546e-08,
+      "loss": 3.8406,
+      "step": 3048960
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 7.132969025061536e-08,
+      "loss": 3.8274,
+      "step": 3049472
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.294374274009525e-08,
+      "loss": 3.8355,
+      "step": 3049984
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 5.455779522957514e-08,
+      "loss": 3.8258,
+      "step": 3050496
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.6171847719055036e-08,
+      "loss": 3.8305,
+      "step": 3051008
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.7802279012266416e-08,
+      "loss": 3.8352,
+      "step": 3051520
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.941633150174631e-08,
+      "loss": 3.8319,
+      "step": 3052032
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.104676279495769e-08,
+      "loss": 3.8331,
+      "step": 3052544
+    },
+    {
+      "epoch": 2.02,
+      "eval_loss": 4.024359226226807,
+      "eval_runtime": 294.2377,
+      "eval_samples_per_second": 1296.88,
+      "eval_steps_per_second": 40.528,
+      "step": 3052726
+    }
+  ],
+  "logging_steps": 512,
+  "max_steps": 3052726,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 10,
+  "total_flos": 1.1757658296514473e+18,
+  "trial_name": null,
+  "trial_params": null
+}